patcher-overview

Stage 4 patch generation playbook. Minimal diffs for validated findings with mandatory patch_verify. Load at patcher-agent startup.

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./packages/decepticon/decepticon/skills/plugins/patcher/SKILL.md

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a tight, highly actionable playbook with excellent workflow sequencing and mandatory validation checkpoints appropriate to destructive patching. Its only weak spot is that the main loop is pseudocode and the bug-class table is slightly dense inline.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence: no concept explanations, tight sections, a compact safe-API table, and minimal prose — every token earns its place.	5 / 5
Actionability	Concrete guidance throughout (patch_propose/patch_verify calls with arguments, a per-bug-class fix table, commit message example), but the core iteration loop is pseudocode control flow rather than copy-paste executable code, leaving minor gaps.	4 / 5
Workflow Clarity	A clearly sequenced multi-step loop with explicit validation (mandatory patch_verify, PoC must fail), feedback loops (revert/retry on tests_failed or regressed), a 3-attempt cap, and a decisive-completion rule — exemplary checkpoint design for a destructive operation.	5 / 5
Progressive Disclosure	Well-organized into clear sections (Iteration loop, Diff style, Commit message format, Decisive-completion rule) with no nested references and no need for bundle files; just over the simple-skill line and the bug-class table is somewhat dense inline, keeping it just short of 5.	4 / 5
	Total	18 / 20 Passed

Description

46%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concrete and distinct within its niche but leans heavily on internal pipeline jargon and lacks an explicit user-facing 'Use when...' trigger clause. It communicates what the skill does but underspecifies when a user would naturally invoke it.

Suggestions

Add an explicit 'Use when...' trigger clause with natural user-facing language (e.g., 'Use when fixing validated security findings and generating minimal patches').

Replace or supplement pipeline jargon ('Stage 4', 'patcher-agent startup') with terms users would actually say, plus common synonyms.

Expand the action list slightly (e.g., mention applying diffs and adding regression tests) to lift specificity toward comprehensive coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ('Stage 4 patch generation playbook') and a couple concrete actions ('Minimal diffs for validated findings', 'mandatory patch_verify'), but coverage is not comprehensive enough for a 4.	3 / 5
Completeness	The 'what' is reasonably clear (generate minimal diffs with mandatory verification), but the 'when' is only a load-time instruction ('Load at patcher-agent startup') with no 'Use when...' trigger clause, capping completeness at 3 per the rubric.	3 / 5
Trigger Term Quality	Keywords like 'patch generation', 'minimal diffs', and 'patch_verify' are internal-pipeline jargon rather than natural phrases a user would say; common synonyms or user-facing trigger terms are missing.	2 / 5
Distinctiveness Conflict Risk	Tied to a highly specific pipeline role ('patcher-agent', 'Stage 4', 'patch_verify'), giving it a clear niche with minimal overlap risk; just short of a 5 because the framing is internal-ecosystem rather than universally distinct.	4 / 5
	Total	12 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: PurpleAILAB/Decepticon
Path: packages/decepticon/decepticon/skills/plugins/patcher/SKILL.md
Commit: 0cf691e

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.