manual_slop/conductor/tracks/superpowers_review_20260619/spec.md

Track Specification: Superpowers Skills Review — Direct Utilization in Manual Slop

Status: Spec approved 2026-06-19 (brainstorming dialogue complete; awaiting user review of written spec). Initialized: 2026-06-19 Owner: Tier 1 Orchestrator (sole author; same pattern as nagent_review_20260608 and fable_review_20260617) Priority: Medium-High (user-explicit; informs future conservative AI-directive improvements) Type: Research-only. No src/ changes. No tests/ changes. No AGENTS.md / conductor/*.md / .opencode/agents/*.md / .opencode/commands/*.md / conductor/code_styleguides/*.md changes. The track produces a reference document for the user's deferred rebuild (parallel to the deferred nagent-rebuild). Format: Conductor convention (per user choice Q4 = A). All artifacts at conductor/tracks/superpowers_review_20260619/.

---

0. Overview

This track produces a critical review of the 14 superpowers-plugin skills against Manual Slop's existing AI-directive corpus and operational practice, with verdicts on which skills are already integrated, which are partially integrated (and where the gaps are), which are not integrated but should be, and which are explicitly not applicable to this project. The deliverable is a reference document the user will use alongside nagent_review_20260608 and fable_review_20260617 when the user eventually rebuilds the project's agent directives.

The review covers all 14 superpowers-plugin skills, plus the project's 5 MMA-tier skills (in a single cluster section), plus the dual-convention problem (docs/superpowers/specs/*.md vs conductor/tracks/<id>/spec.md) that the user explicitly flagged. The verdict taxonomy is hybrid: a primary verdict (nagent-style: PARITY / PARTIAL / GAP / ARCH-DIFF / SUBSUMED) plus a secondary integration tag (INTEGRATED / INTEGRATE-PARTIAL / INTEGRATE / REJECT-WITH-REASON / N/A).

The track is research-only. No src/ files are modified. No agent-directive files (AGENTS.md, conductor/*.md, .opencode/agents/*.md, .opencode/commands/*.md, conductor/code_styleguides/*.md) are modified. The actual conservative changes become follow-up tracks in the user's deferred rebuild.

The user's framing (2026-06-19): "conservative changes incrementally to improve AI performance and quality standards of output. I'm not after speed, pure discipline, high grade inference, good tool use, and careful text generation." The review's lens is AI quality (discipline + inference + tool use + text generation), not AI speed.

---

1. Current State Audit (as of commit f0f404632)

1.1 Already Implemented (DO NOT re-implement)

What	Where	Notes
The project's agent-directive corpus (the target the review evaluates against)	AGENTS.md (root, 200 lines); conductor/*.md (7 files, ~3,000 lines); conductor/code_styleguides/*.md (11 files, ~2,400 lines); .opencode/agents/*.md (6 files, ~1,100 lines); .opencode/commands/*.md (9 files, ~700 lines); docs/*.md excluding superpowers/ (~16,000 lines across 40+ files including 36 guide_*.md)	The review reads this corpus; it does not modify it.
The superpowers plugin content (the subject of the review)	C:\Users\Ed\.cache\opencode\packages\superpowers@git+https_\github.com\obra\superpowers.git\node_modules\superpowers\skills\	14 skills, each with a SKILL.md. Read at the start of the review.
The project's 5 MMA-tier skills (the local comparison)	.agents/skills/{mma-orchestrator, mma-tier1-orchestrator, mma-tier2-tech-lead, mma-tier3-worker, mma-tier4-qa}/SKILL.md	Mirrored at .gemini/skills/ (legacy; left over from the Gemini CLI conductor-plugin era; should be re-evaluated in the deferred rebuild).
The chronology track (the immediate predecessor)	conductor/tracks/chronology_20260619/	This track is blocked_by chronology_20260619 per user directive.
The nagent_review corpus (the primary precedent)	conductor/tracks/nagent_review_20260608/	11 files; 4,969-line v2.3 rewrite is the template for this track's structure. The verdict taxonomy borrows PARITY / PARTIAL / GAP / ARCH-DIFF / SUBSUMED from this corpus.
The fable_review corpus (the secondary precedent)	conductor/tracks/fable_review_20260617/	The cluster + synthesis pattern from this corpus is not used here (the superpowers review is smaller and single-author); but the "things I notice that don't fit the main sections" pattern (Section 16) is borrowed.
The intent_dsl_survey (the sibling reference)	conductor/tracks/intent_dsl_survey_20260612/	The user explicitly named this as a sibling. The bridge artifact (nagent_takeaways_superpowers_20260619.md) parallels this track's relation to nagent_review.
The dual-convention situation (the user-flagged finding)	docs/superpowers/specs/ (20 files) + docs/superpowers/plans/ (21 files) co-exist with conductor/tracks/<id>/spec.md + plan.md	The OLD convention is conductor/tracks/<id>/ (started when Gemini CLI was actively used with the conductor plugin); the NEW convention is docs/superpowers/specs/ + docs/superpowers/plans/ (per superpowers-plugin defaults). Section 16 of the review analyzes the situation.

1.2 Gaps to Fill (This Track's Scope)

• The synthesis report (report.md, 16 sections). Does not exist. Will be authored by Tier 1 across 21 atomic commits.
• The 20-row comparison table (comparison_table.md). Does not exist. Flat reference: one row per superpowers skill × verdict × recommendation.
• The decisions file (decisions.md, ~15-25 entries). Does not exist. Sorted by priority; each entry has a "destination file" field so the user can batch the deferred rebuild.
• The nagent_takeaways bridge (nagent_takeaways_superpowers_20260619.md, ~150 lines). Does not exist. Links this track's findings to nagent_takeaways_20260608.md and fable_review_20260617/report.md so the user can read all three reviews as a unified corpus.

1.3 Pre-Existing Conditions the Track Must Respect

• Chronology is current_phase=0 and not yet started. The Phase 8 cross-check (165+ rows of conductor/chronology.md) is the dominant scope; this track cannot start until chronology ships.
• The project's TDD / verification-before-completion discipline (per AGENTS.md "Critical Anti-Patterns") is already close to the superpowers-plugin's test-driven-development + verification-before-completion skills. The review's verdicts will reflect this (likely PARITY or INTEGRATED-PARTIAL for both).
• The .opencode/agents/ and .opencode/commands/ configurations (Gemini CLI era) are not used by OpenCode; they're leftover from the conductor-plugin era. Section 16 will flag this.
• The data-oriented error handling convention (per conductor/code_styleguides/error_handling.md) is philosophically aligned with the superpowers-plugin's systematic-debugging skill's "root cause before fix" stance; the review surfaces this alignment.
• The nagent_review's deferred rebuild (per conductor/tracks/nagent_review_20260608/spec.md §10) is the next major agent-directive overhaul the user has queued. This track's recommendations are additional inputs to that rebuild, not a competing one.

---

2. Goals (Priority Order)

Priority	Goal	Rationale
A (primary)	The synthesis report (report.md, 16 sections) covers all 14 superpowers-plugin skills + the 5 MMA skills cluster + the dual-convention + anything else cross-cutting findings.	The report is the deliverable.
A (primary)	Every section ends with a hybrid verdict block (primary nagent-style + secondary integration tag).	The verdict block is the unit of actionability. The user uses the verdicts to plan the deferred rebuild.
A (primary)	The 20-row comparison_table.md is the at-a-glance reference; the decisions.md is the prioritized rebuild backlog.	The two artifacts are how the user consumes the review at scale.
B (analytical)	The "anything else" findings in Section 16 are bounded (one paragraph each) and don't balloon into a full re-review.	Scope discipline; bigger findings become follow-up tracks.
B (process)	The nagent_takeaways_superpowers_20260619.md bridge points to the relevant sections of nagent_review_20260608 and fable_review_20260617 for cross-reference.	The user wants to read all three reviews as a unified corpus.
B (process)	The verdict block template is consistent across all 16 sections (same fields, same vocabulary).	The self-review pass (Phase 8) is the check.
C (housekeeping)	conductor/tracks.md is updated to register the track in the appropriate section.	Standard per-track convention.
C (housekeeping)	The 21 commits are atomic with git notes attached per the project's convention.	conductor/workflow.md §"Task Workflow" step 9.2.

---

3. Functional Requirements

3.1 The 16 Sections of report.md

#	Section	Skill/topic	Depth
1	Using Superpowers	using-superpowers	Brief (50-100 LOC)
2	Brainstorming	brainstorming	Deep-dive (200-400 LOC)
3	Writing Plans	writing-plans	Deep-dive (200-400 LOC)
4	Test-Driven Development	test-driven-development	Deep-dive (200-400 LOC)
5	Verification Before Completion	verification-before-completion	Deep-dive (200-400 LOC)
6	Systematic Debugging	systematic-debugging	Deep-dive (200-400 LOC)
7	Subagent-Driven Development	subagent-driven-development	Deep-dive (200-400 LOC)
8	Executing Plans	executing-plans	Medium (100-250 LOC)
9	Dispatching Parallel Agents	dispatching-parallel-agents	Brief (50-150 LOC)
10	Receiving Code Review	receiving-code-review	Medium (100-250 LOC)
11	Requesting Code Review	requesting-code-review	Brief (50-150 LOC)
12	Finishing a Development Branch	finishing-a-development-branch	Brief (50-150 LOC)
13	Using Git Worktrees	using-git-worktrees	Brief (50-150 LOC)
14	Writing Skills	writing-skills	Medium (100-250 LOC)
15	MMA Skills Cluster	All 5 project MMA skills	Cluster (300-500 LOC; 5 sub-sections, each with its own verdict block)
16	Dual-Convention + Anything Else	Cross-cutting	Medium (200-400 LOC; one paragraph per finding)

Total report scope: ~2,800-4,500 LOC across 16 sections. ~280 LOC average per section.

3.2 The Verdict Block Template (per section)

Every section ends with this block (verbatim):

**Verdict.**

| Field | Value |
|---|---|
| **Primary** | `<PARITY | PARTIAL | GAP | ARCH-DIFF | SUBSUMED>` |
| **Integration tag** | `<INTEGRATED | INTEGRATE-PARTIAL | INTEGRATE | REJECT-WITH-REASON | N/A>` |
| **Section size** | `<brief | medium | deep-dive | cluster>` |
| **Cross-refs** | `<nagent_review_20260608 §X.Y, fable_review_20260617 §X.Y, intent_dsl_survey_20260612 §X.Y>` (if any; "none" if N/A) |

**Rationale.** [1-3 sentences.]

**Recommended change.** [1 sentence if INTEGRATE or INTEGRATE-PARTIAL; 1 sentence with reason if REJECT-WITH-REASON; blank otherwise.]

Verdict vocabulary (locked):

Primary	Definition
PARITY	Manual Slop already applies this skill fully. Nothing to do.
PARTIAL	Manual Slop applies this skill with documented gaps. The gaps are the recommended change.
GAP	Manual Slop does not apply this skill, and should. The full skill integration is the recommended change.
ARCH-DIFF	The skill's design doesn't fit Manual Slop's architecture. Don't force-fit; flag the architectural mismatch in the rationale.
SUBSUMED	The skill's purpose is achieved by another Manual Slop mechanism (e.g., the project's 4-tier MMA subsumes nagent's --description self-describing-executables pattern). Cite the subsuming mechanism.

Integration tag	Definition
INTEGRATED	Already in place. The user can re-affirm in the deferred rebuild without code change.
INTEGRATE-PARTIAL	Apply the skill where the gaps are. The "Recommended change" sentence specifies which gaps.
INTEGRATE	Add the skill (or a Manual Slop-specific adaptation of it) to the agent directives.
REJECT-WITH-REASON	Do not integrate. The "Recommended change" sentence is a reason (not a "do nothing").
N/A	The skill does not apply to Manual Slop's domain (Application + Meta-Tooling).

3.3 The comparison_table.md Format

20-row table. Columns:

Skill	Primary verdict	Integration tag	Section LOC	Recommended change	Cross-ref

Where:

• Skill = one of: 14 superpowers-plugin skills, 5 MMA skills (one row each), or "Dual-Convention + Anything Else" (one row).
• Cross-ref = the relevant sections of nagent_review_20260608 and fable_review_20260617 (or "none").

3.4 The decisions.md Format

~15-25 entries, sorted by priority (HIGH → MEDIUM → LOW). Each entry:

Field	Value
#	Sequential ID
Priority	HIGH / MEDIUM / LOW
Skill	Which superpowers skill this is for
Change	1-sentence description of the conservative change
Destination file	Where the change goes in the deferred rebuild (e.g., "AGENTS.md §Critical Anti-Patterns", "new conductor/code_styleguides/superpowers_integration.md", "new .agents/skills/superpowers-bridge/SKILL.md")
Effort	S / M / L / XL (per conductor/workflow.md Tier 1 rules — no day estimates)
Evidence	report.md §N + verdict block quote

Empty-cell rule: if the "Change" cell is empty, the entry is PARITY / INTEGRATED / N/A and the deferred rebuild doesn't need to do anything. Empty cells = no rebuild action.

3.5 The nagent_takeaways_superpowers_20260619.md Bridge

~150 LOC. Format:

1. TL;DR (1 paragraph): "This bridge connects the superpowers review's verdicts to the nagent_review's 16 future-track candidates. The two corpora overlap on X, diverge on Y, and the superpowers review adds Z new candidates."
2. Cross-reference table (~10-15 rows): one row per superpowers verdict that touches an nagent candidate, columns: superpowers section | verdict | nagent candidate | relationship (subsumes / extends / contradicts / independent).
3. The 3 new candidates the superpowers review adds (not in nagent_review): one paragraph each, with verdict evidence.
4. The 2 nagent candidates the superpowers review contradicts (if any): one paragraph each, with verdict evidence.
5. Pointer to fable_review (1 paragraph): which fable_review sections the user should read alongside which superpowers sections.

---

4. Non-Functional Requirements

4.1 Process Discipline

• All 21 commits are atomic (per conductor/workflow.md §"Task Workflow" step 9).
• Every commit has a git note attached (per step 9.2) summarizing the section.
• All tasks are recorded in state.toml with commit SHAs.
• No day / hour / minute estimates in any track artifact. T-shirt size only.
• The 1-space indentation rule applies to metadata.json and state.toml (the only Python-shaped files). Markdown is not Python; the rule doesn't apply to prose.
• The "no diagnostic noise in production" rule doesn't apply (no src/ changes).
• The "HARD BAN: git restore / git checkout -- <file> / git reset" rule applies per AGENTS.md.
• No new src/<thing>.py files (per AGENTS.md "File Size and Naming Convention" hard rule).
• No new scripts/audit_*.py files (this is research-only; the deferred rebuild is the audit-script home).

4.2 Documentation Conventions

• The synthesis report uses the 1-sentence-per-line pattern for dense content (per conductor/product-guidelines.md §"AI-Optimized Compact Style").
• The synthesis report uses tables for the verdict block (per §3.2 above).
• All file:line references in the synthesis report are stable (the report is the durable artifact; the superpowers-plugin source may evolve).

4.3 Audit Hooks

This track is research-only; no scripts/audit_*.py scripts are added or modified. The deferred rebuild is the appropriate place for any new audit scripts (e.g., a "dual-convention auditor" that flags any new spec.md file appearing outside conductor/tracks/<id>/).

---

5. Architecture Reference

• conductor/tracks/nagent_review_20260608/ — the primary precedent. The verdict taxonomy (PARITY / PARTIAL / GAP / ARCH-DIFF / SUBSUMED) is borrowed from report.md §0.2. The "one section per pattern" structure is borrowed from §2.
• conductor/tracks/fable_review_20260617/ — the secondary precedent. The "anything else" cross-cutting findings pattern (Section 16) is borrowed from §2 ("In dialogue with the intent DSL survey"). The cluster-sub-agent dispatch pattern is not used (single-author is simpler for the smaller corpus).
• conductor/tracks/intent_dsl_survey_20260612/ — the sibling reference track. The user named this as a sibling; the bridge artifact (nagent_takeaways_superpowers_20260619.md) parallels this track's relation to nagent_review.
• conductor/tracks/chronology_20260619/ — the immediate predecessor. This track is blocked_by chronology_20260619 per user directive (2026-06-19).
• AGENTS.md (root, 200 lines) — the project's top-level agent-facing rules. Sections 4-7 (TDD, verification, debugging, subagent-driven development) reference this file.
• conductor/workflow.md (63K) — the operational workflow. Sections 3, 4, 5, 6 (writing-plans, TDD, verification, debugging) reference the TDD protocol + Process Anti-Patterns.
• conductor/code_styleguides/ (11 files, ~140K) — the convention catalog. Section 16 (dual-convention + anything else) and the MMA cluster (Section 15) reference these.
• .opencode/agents/*.md (6 files) — the 4 MMA tier agents + explore + general. Section 15 (MMA cluster) reads these. Note: the .opencode/ directory is a legacy from the Gemini CLI conductor-plugin era and is not used by OpenCode; the project's actual MMA skills live in .agents/skills/. The mirror at .gemini/skills/ is similarly legacy. Section 16 flags this.
• .agents/skills/*.md (5 files) — the project's current MMA-tier skills (the local comparison in Section 15).
• docs/AGENTS.md — the agent-facing mirror of docs/Readme.md. Section 16 references this.
• docs/guide_*.md (36 files, ~580K) — the 14 deep-dive guides. Sections 7, 8, 15 reference these selectively.
• Superpowers plugin content — C:\Users\Ed\.cache\opencode\packages\superpowers@git+https_\github.com\obra\superpowers.git\node_modules\superpowers\skills\. 14 skills; each has a SKILL.md. The subject of the review.
• docs/superpowers/specs/ (20 files) + docs/superpowers/plans/ (21 files) — the NEW convention. Section 16 analyzes the dual-convention situation.

---

6. Implementation Phases (10 phases, 21 commits)

#	Phase	Scope	Commits
1	Setup	Create track directory. Write skeleton files (this spec.md, metadata.json, state.toml with current_phase=1, report.md with 16 section headers + empty bodies, comparison_table.md with column headers, decisions.md with template, nagent_takeaways_superpowers_20260619.md empty). Update conductor/tracks.md "Active" section to register the track.	1
2	Sections 1-4 (1 brief + 3 deep-dives)	using-superpowers, brainstorming, writing-plans, test-driven-development.	4
3	Sections 5-8 (3 deep-dives + 1 medium)	verification-before-completion, systematic-debugging, subagent-driven-development, executing-plans.	4
4	Sections 9-14 (2 brief + 2 medium + 2 brief)	dispatching-parallel-agents, receiving-code-review, requesting-code-review, finishing-a-development-branch, using-git-worktrees, writing-skills.	6
5	Section 15 (MMA cluster)	5 sub-sections: mma-orchestrator, mma-tier1-orchestrator, mma-tier2-tech-lead, mma-tier3-worker, mma-tier4-qa. Each with verdict block.	1
6	Section 16 (cross-cutting)	Dual-convention analysis + "anything else" findings (one paragraph each).	1
7	Side artifacts	comparison_table.md (20 rows), decisions.md (~15-25 entries), nagent_takeaways_superpowers_20260619.md (bridge).	3
8	Self-review	Per the brainstorming skill: placeholder scan, internal consistency, scope check, ambiguity check. Fix inline.	0
9	User review	User reviews report.md + side artifacts. Approves or iterates.	0
10	Finalize	Update state.toml to current_phase=10. Register track as "Recently Completed" in conductor/tracks.md. Update metadata.json with final statistics (commit count, LOC, verdict distribution).	1

Total commits: 1 + 4 + 4 + 6 + 1 + 1 + 3 + 1 = 21 atomic commits.

---

7. Verification Criteria

The track is "done" when all of the following are true:

• report.md has all 16 sections present and non-empty.
• Every section ends with the hybrid verdict block (per §3.2).
• comparison_table.md has all 20 rows (14 superpowers + 5 MMA + 1 dual-convention).
• decisions.md has 15-25 entries, sorted by priority (HIGH → MEDIUM → LOW), with empty cells for PARITY / INTEGRATED / N/A verdicts.
• nagent_takeaways_superpowers_20260619.md exists with the 5-part bridge structure (TL;DR + cross-reference table + new candidates + contradictions + fable pointer).
• No src/ / tests/ / AGENTS.md / conductor/*.md / .opencode/agents/*.md / .opencode/commands/*.md / conductor/code_styleguides/*.md changes (research-only).
• Self-review pass complete (placeholder scan, internal consistency, scope check, ambiguity check).
• User has reviewed and approved the final report + side artifacts.
• conductor/tracks.md updated to register the track.
• All 21 commits are atomic with git notes attached.
• state.toml final state is current_phase=10 and status="active" (until archived per the chronology track's archive convention).
• No new src/*.py or scripts/audit_*.py files created (per AGENTS.md hard rules).

---

8. Risks & Mitigations

Risk	Impact	Likelihood	Mitigation
Section verdict inconsistency (some sections use PARITY, others use GAP for the same condition)	Medium (the comparison_table.md becomes hard to scan)	Medium	The verdict block template (§3.2) is fixed; the self-review pass (Phase 8) catches inconsistencies.
The "anything else" findings in Section 16 balloon into a full re-review of the codebase	Medium (scope creep)	Medium	Section 16 has a hard limit: findings are one paragraph each. Anything bigger becomes a follow-up track and is logged in decisions.md.
decisions.md becomes a wish-list rather than prioritized conservative changes	Low (the user reviews before approving)	Medium	The user-review gate (Phase 9) is the check. The decisions.md format requires a "Destination file" field so the user can spot scope-creep recommendations.
nagent_takeaways_superpowers_20260619.md bridge is too thin	Low (it's a small artifact)	Low	The bridge is intentionally ~150 LOC; it's a pointer, not a co-equal report.
The 21 commits become hard to review (user has to read 21 git notes)	Low (atomic commits are the project's convention)	Low	The commits are mechanical; the user reviews the report as a single document, not the commit-by-commit progression.
The dual-convention section (16) argues for a position the user disagrees with	Low (user-review gate catches it)	Medium	The section presents both options (keep conductor convention vs. adopt superpowers convention vs. split by artifact type); the user picks in the deferred rebuild.
Chronology track takes longer than expected and delays this track	Low (no impact on this track's quality)	High	This track is blocked_by chronology_20260619; the order is fixed. The chronology track is on its own clock.
The superpowers plugin updates between the start of the review and the end	Low (the report is a snapshot)	Low	The report notes the plugin version / commit at the start of Phase 2 and is dated 2026-06-19. If the plugin updates mid-review, the report flags the version mismatch in the verdict rationale.

---

9. Out of Scope (Explicit)

1. Modifying any agent-directive file in the project. The recommendations go in decisions.md for the deferred rebuild.
2. Building any recommendation. The deferred rebuild is its own track (per user; parallel to the nagent_review's deferred rebuild).
3. Reviewing every external AI corpus (nagent, Fable, Claude, OpenAI, etc.). The superpowers plugin is the named subject; the project's MMA skills are the local comparison; everything else is referenced only when directly relevant.
4. Doing a "review of all 14 skills in equal depth." Some skills (e.g., using-superpowers, using-git-worktrees) are foundational and get a brief verdict; some (e.g., brainstorming, test-driven-development, writing-plans) get full deep-dives because they shape every track the project runs.
5. Rewriting or migrating docs/superpowers/specs/*.md → conductor/tracks/<id>/spec.md. The dual-convention analysis is in Section 16; the migration (if any) is the deferred rebuild's work.
6. Adding new .opencode/agents/*.md files, new conductor/code_styleguides/*.md files, or new scripts/audit_*.py scripts. The report may recommend these; the rebuild creates them.
7. Running automated tests. The track is research-only; verification is the brainstorming-skill self-review plus user review.
8. Creating new docs/Readme.md or docs/AGENTS.md entries. The report is at conductor/tracks/superpowers_review_20260619/; it is not in the docs index.
9. The user's deferred nagent-rebuild itself. The recommendations in decisions.md are additional inputs to that future track; the rebuild is not this track.

---

10. See Also

10.1 Internal References

• conductor/tracks/chronology_20260619/ — the immediate predecessor. This track is blocked_by it.
• conductor/tracks/nagent_review_20260608/ — the primary precedent. Verdict taxonomy + section structure are borrowed from here.
• conductor/tracks/fable_review_20260617/ — the secondary precedent. The "anything else" cross-cutting findings pattern is borrowed from here.
• conductor/tracks/intent_dsl_survey_20260612/ — the sibling reference track. The bridge artifact parallels this track's relation to nagent_review.
• AGENTS.md (root) — the project's top-level agent-facing rules. Sections 4-7 reference this.
• conductor/workflow.md — the operational workflow. Sections 3-6 reference the TDD protocol + Process Anti-Patterns.
• conductor/product.md — the product vision. Section 15 (MMA cluster) and Section 16 reference the 4-tier MMA description.
• conductor/product-guidelines.md — the AI-Optimized Compact Style. Sections 2, 5, 7 reference the formatting heuristics.
• conductor/tech-stack.md — the tech stack. Section 16 references the tools inventory + provider list.
• conductor/code_styleguides/ (11 files) — the convention catalog. Section 15 references these; Section 16 flags any missing conventions.
• .agents/skills/*.md (5 files) — the project's current MMA-tier skills. Section 15 reads these.
• .opencode/agents/*.md (6 files) — the legacy Gemini CLI conductor-plugin files. Section 16 flags these as legacy.
• docs/AGENTS.md — the agent-facing mirror. Section 16 references this.
• docs/guide_*.md (36 files) — the 14 deep-dive guides. Sections 7, 8, 15 reference these selectively.
• docs/superpowers/specs/ (20 files) + docs/superpowers/plans/ (21 files) — the NEW convention. Section 16 analyzes the dual-convention situation.
• Superpowers plugin content — C:\Users\Ed\.cache\opencode\packages\superpowers@git+https_\github.com\obra\superpowers.git\node_modules\superpowers\skills\. 14 skills. The subject of the review.

10.2 External References

• The superpowers plugin: https://github.com/obra/superpowers (the source of all 14 skills). The plugin's using-superpowers skill is the project's "always start here" reference.
• Mike Acton's nagent: https://github.com/macton/nagent (the source of the nagent_review corpus; this track borrows the verdict taxonomy from report.md).
• Anthropic's Claude Fable: docs/artifacts/Fable System Prompt.txt (local-only; the source of the fable_review corpus; this track's Section 16 cross-references the fable review's relevant sections).

10.3 Track-internal References

• conductor/tracks/superpowers_review_20260619/spec.md — this file.
• conductor/tracks/superpowers_review_20260619/metadata.json — the track metadata (id, scope, blocks, etc.).
• conductor/tracks/superpowers_review_20260619/state.toml — the track state (current_phase, task tracking).
• conductor/tracks/superpowers_review_20260619/report.md — the main 16-section synthesis report (executed by Tier 1 in Phases 2-6).
• conductor/tracks/superpowers_review_20260619/comparison_table.md — the 20-row flat reference (executed by Tier 1 in Phase 7).
• conductor/tracks/superpowers_review_20260619/decisions.md — the prioritized rebuild backlog (executed by Tier 1 in Phase 7).
• conductor/tracks/superpowers_review_20260619/nagent_takeaways_superpowers_20260619.md — the bridge to nagent_review + fable_review (executed by Tier 1 in Phase 7).