AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

H4tcher Capability Remove

skill-hatch3r-hatch3r-h4tcher-capability-remove · by hatch3r

Decommission a hatch3r artifact (agent, skill, rule, command, hook, adapter, pipeline module) after passing the D16.3 removal-threshold gate, drafting migration notes, and scrubbing every cross-reference.

No reviews yet
0 installs
34 views
0.0% view→install

Install

$ agentstack add skill-hatch3r-hatch3r-h4tcher-capability-remove

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-hatch3r-hatch3r-h4tcher-capability-remove)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of H4tcher Capability Remove? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

> Last updated: 2026-06-06

Capability Remove (Maintainer)

Maintainer preset for phasing out a hatch3r artifact. Hardest preset to invoke: a D16.3 gate blocks removal unless three criteria all hold, and migration notes plus reference scrubbing run before the maintainer is allowed to commit.

§0.1: Ambiguity Gate (P8 B1)

Per CONSTITUTION §2 P8 B1: every framework-dev workflow mutating canonical artifacts detects and resolves ambiguity via agents/shared/user-question-protocol.md BEFORE executing — default behavior, not exception-driven. Apply the protocol when any hold: (a) ambiguous scope (which artifact, and remove vs merge-then-deprecate per D16.3 add-vs-remove bias), (b) multiple valid interpretations with materially different migration cost, (c) irreversible action — removal is inherently destructive, so confirm the target id and that no consumer depends on it, or (d) missing acceptance criteria (no migration-note target, no definition of done). Use the platform-native question tool; one question per turn; 2-4 numbered options with one-line trade-offs; declare the default-if-no-response. Do NOT proceed to Step 0 until resolved or the default is confirmed.

Step 0: Triage

Classify the removal target before any work runs.

| Tier | Criteria | Pipeline | |------|----------|----------| | T1 | Leaf artifact with 0 inbound references | Inline; skip Step 4; full Step 6 gates | | T2 | Artifact with 1-3 references requiring migration notes | 1-3 parallel sub-agents; full gates | | T3 | Artifact named in any agentPipeline: array, listed in governance/inventory.json registry, or referenced from governance/CONSTITUTION.md / governance/VISION.md | >=3 parallel sub-agents; full gates + CL-3 flag if D16/AUDIT touched |

Record tier + rationale at the top of the workspace summary.

Step 1: Preflight

  1. Confirm clean tree + branch: git status --short empty; git branch --show-current matches an open release/feature branch.
  2. Cache lean-threshold limits from governance/CONSTITUTION.md §2 P5.
  3. Identify the target artifact type (agent | skill | rule | command | hook | adapter | pipeline module) and its canonical path.
  4. Capture baseline wc -l on governance/inventory.json and any governance file likely to be edited.

Step 2: Removal-Qualification Gate (D16.3)

Run the D16.3 removal-threshold check inline (governance/audit/domains/D16-compound-system.md §16.3 "Removal candidate threshold"). Removal qualifies only when ALL three criteria hold:

  • (a) Zero unique value beyond an existing artifact (run duplication scan; cite the artifact that subsumes the candidate).
  • (b) " agents/ skills/ rules/ commands/ hooks/ checks/ prompts/ github-agents/ src/ governance/ docs/ docs-site/ tests/ CLAUDE.md README.md CHANGELOG.md`; record the count).
  • (c) No orchestrator dependency in any commands/hatch3r-*.md agentPipeline: array (grep -l "" commands/hatch3r-*.md; inspect each match).

Cite which criterion is crossed for every failing check. Block immediately on any failure and recommend /h4tcher-capability-refactor (merge) per the D16.3 add-vs-remove bias check.

Step 3: Web Research

Skipped by default. Run only when the removal cites an external "industry has deprecated this" claim. When triggered, follow the rigor contract (agents/shared/rigor-contract.md §"Web Research Mandate"): >=2 independent sources, trust tier (official-docs > peer-reviewed > vendor-note > independent-analysis > blog-post), recency window (=3 on T3 across the migration-notes / refs-canonical / refs-code-docs slots above. Token cost never serializes independent work (.claude/rules/fan-out-discipline.md Cost-dominance clause). Emit sub_agents_spawned: { count, rationale } in your output (the Sub-agents field of the Step 7 summary).

Step 5: Cross-Skill Delegation

None for body authoring during removal. The migration-notes slot may call /h4tcher-content-author only when a brand-new replacement artifact must be authored — rare; document the rationale in the workspace summary when invoked.

Step 6: Governance Gates

Run the full hatch3r gate battery PLUS removal-specific checks. Block commit on any failure.

| Gate | Command | Threshold | |------|---------|-----------| | Validate | npm run validate | 0 errors | | Tests | npm test | 0 failed | | Typecheck | npx tsc --noEmit | 0 errors | | Lint | npm run lint | 0 errors | | Build | npm run build | succeeds; dist/ produced | | Rule parity | npm run validate:rule-parity | 0 mismatches | | Efficiency invariants | npm run validate:efficiency | 0 violations | | Inventory regen | npm run inventory then git diff governance/inventory.json | diff matches the removal exactly; no unexpected drift | | Inventory docs drift | npm run inventory:check-docs | 0 drift | | Reference-orphan check | grep -rn "" agents/ skills/ rules/ commands/ hooks/ checks/ prompts/ github-agents/ src/ governance/ docs/ docs-site/ tests/ CLAUDE.md README.md CHANGELOG.md | 0 hits outside the CHANGELOG migration entry | | CHANGELOG entry | grep -n "" CHANGELOG.md | >=1 line under the active release's ### Removed section | | Anti-slop | grep -nE against .claude/rules/anti-slop-enforcement.md wordlist on every modified .md / .mdc | 0 hits | | Pillar compliance | Apply .claude/rules/pillar-compliance.md to the removal | Removal serves one or more of P1-P8 | | Lean thresholds | wc -l on every modified governance file | Within CONSTITUTION §2 P5 limits |

Step 7: Stop-Before-Commit Summary

Emit the slot-filled template below. Do not commit, branch, push, or merge.

Capability-Remove Summary
-------------------------
Artifact:           
Type:               agent | skill | rule | command | hook | adapter | pipeline-module
Tier:               T1 | T2 | T3
D16.3 criteria:     (a)  (b)  (c) 

Removed paths:      
Migration notes:    CHANGELOG.md (line , ### Removed) [+ docs-site/.md if authored]
agentPipeline diffs:  after, or "none">
Inventory delta:     after in governance/inventory.json>

Gates:
  npm run validate                
  npm test                        
  npx tsc --noEmit                
  npm run lint                    
  npm run build                   
  npm run validate:rule-parity    
  npm run validate:efficiency     
  npm run inventory:check-docs    
  Reference-orphan grep           
  CHANGELOG entry present         
  Anti-slop hits                  

Pillars served:     P, P
Sub-agents:         count==3>, rationale=
Sources (web):      
Confidence:         high | medium | low

Next action (maintainer runs manually):
  git add 
  git commit -s -m "refactor(): remove "
  git push

Constraints

  • Block on any D16.3 failure in Step 2; do not proceed without the maintainer's explicit override + recorded rationale.
  • Never hand-edit governance/inventory.json — only via npm run inventory.
  • Do not skip the reference-orphan grep; one orphan hit blocks commit.
  • Do not commit, branch, push, or merge from inside this skill.
  • DCO sign-off via git commit -s is required when the maintainer commits.

References

  • D16.3 removal-threshold + add-vs-remove bias check: governance/audit/domains/D16-compound-system.md §16.3 "Removal candidate threshold"
  • Web research mandate: agents/shared/rigor-contract.md
  • Quality charter: agents/shared/quality-charter.md
  • Sibling presets: .claude/skills/h4tcher-capability-discover/SKILL.md, .claude/skills/h4tcher-capability-refactor/SKILL.md
  • Author skill (replacement artifact, rare): .claude/skills/h4tcher-content-author/SKILL.md

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.