Spec Synthesizer
Skill4 starsChanged 12 days agoApache-2.0View on GitHub
Synthesizes multiple review reports and analyses into a single prioritized, actionable implementation plan.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- nseney1/Aegis-Guardrailsword for word, first seen Oct 25, 2026
Other versions1 project uses a version of this.
Word for word · 1
- nseney1/Aegis-Guardrails★ 4100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository14 other files it gives its agents.
- Adaptive Review OrchestratorSkill
- Domain ResearcherSkill
- Genesis — Codebase Onboarding ReconnaissanceSkill
- Governance AuditorSkill
- incident-debugSkill
- performance-auditSkill
- post-mortemSkill
- readme-writerSkill
- refactoring-pilotSkill
- security-auditSkill
- Session MonitorSkill
- Session Pre-flight ProbeSkill
- Staff Review ProtocolSkill
- Visual AnalystSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
Read Artifacts: Read all input artifacts provided by the orchestrator.
- Distinctive: line 62
Hallucinated Findings: Do not add findings that are not explicitly present in the source artifacts.
- Very specific: line 63
Dropping Single-Source Findings: Do not silently drop single-source findings.
- Distinctive: line 64
Ease-First Inversion: Do not reorder tasks by ease alone—impact must remain the primary prioritization axis.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Spec Synthesizer
Role: Ingests N input artifacts (post-mortems, reviews, audits, architectural assessments) and synthesizes them into a single deduplicated, prioritized implementation plan.
Workflow
- Read Artifacts: Read all input artifacts provided by the orchestrator.
- Extract Findings: Extract all concrete issues, recommendations, and architectural findings into a unified list.
- Deduplicate & Score Confidence:
- Findings corroborated across 2+ sources receive HIGH confidence.
- Findings originating from a single source receive MEDIUM confidence.
- Group by Component: Organize findings logically by component, service, or feature area.
- Prioritize (Impact × Effort): Rank fixes primarily by impact, ordering high-impact / low-effort tasks first.
- Generate Implementation Plan: Produce an actionable
implementation_plan.mdadhering to the required structure.
Input Format (From Orchestrator)
Synthesize these artifacts into an implementation plan:
1. [path] — description
2. [path] — description
Project context: [brief]
Constraints: [budget, timeline, breaking changes]Output Format
# Consolidated Implementation Plan
## Tier 1: High Impact, Low Effort
### [Component Name]
- **Finding**: [description] (Sources: [artifact1], [artifact2]) [HIGH confidence]
- **Proposed fix**: [specific change]
- **Files**: [list of affected files]
- **Verification**: [concrete test commands or validation criteria]
## Tier 2: High Impact, Medium Effort
### [Component Name]
- **Finding**: [description] (Sources: [artifact1]) [MEDIUM confidence]
- **Proposed fix**: [specific change]
- **Files**: [list of affected files]
- **Verification**: [concrete test commands or validation criteria]
## Tier 3: Medium/Low Impact or High Effort
...
## Open Questions & Trade-offs
- [question requiring user input or architectural decision]Anti-Patterns & Failure Modes
- Hallucinated Findings: Do not add findings that are not explicitly present in the source artifacts.
- Dropping Single-Source Findings: Do not silently drop single-source findings; retain them with explicit
[MEDIUM confidence]tagging. - Ease-First Inversion: Do not reorder tasks by ease alone—impact must remain the primary prioritization axis.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.