mortennordbye/agentic-dr/fix
Claude Code and agent skill by mortennordbye/agentic-dr
- community
Checking that the SKILL.md file is still in the repository.
Adaptive remediation of ONE disaster-recovery service that will not come up — a region-parity gap or a misbehaving managed service (ARCHITECTURE.md Mode 3). Diagnoses via the dr-dynamic-remediator agent, proposes only invariant-safe alternatives, escalates rather than weaken isolation, and on approval implements a surgical fix and re-plans through the failover gates. Use when asked to fix, remediate or work around one failing service in the DR estate.
Install
npx vskill@latest install mortennordbye/agentic-dr/fixSKILL.md
/agentic-dr:fix <service> — adaptive remediation of one service (Mode 3)
Not regeneration — problem-solving against one failing service (a region-parity gap or a misbehaving managed service). Reached deliberately, or escalated here by /agentic-dr:failover Phase-5 triage (route_to: human / escalate_to_mode3).
Do not run a whole build to get here. If the service has no generated DR root yet, that is a /agentic-dr:dry-run job first.
The rules that still hold
- NEVER run `gh workflow run` — including `plan_only=true` — without explicit user approval.
Stop and ask every time (CLAUDE.md, ARCHITECTURE §10).
- DR estate only. Read the source roots
profile/repo-map.mdnames as immutable truth; write
only under the DR output tree and state dir it names; trigger only the DR pipeline it names (ARCHITECTURE §1).
- The customer profile is authoritative for every concrete value — never memory.
Procedure
Spawn the `agentic-dr:dr-dynamic-remediator` agent with the failure + the affected DR root. It diagnoses, then proposes only invariant-safe alternatives:
Hard invariants it may never propose violating (ARCHITECTURE §3): no public exposure of aninternal service, no egress bypass of the controlled path, no weakening of network isolation /encryption-at-rest / TLS, and DR-estate-only. If the only working option breaks one, it **stops andescalates to a human** instead of presenting it.
[APPROVAL GATE] the engineer picks an option → implement it as a surgical change to the affected DR root (or record a manual step) → re-plan/apply through the /agentic-dr:failover gates. An approved + applied fix auto-drafts a PR to profile/region-gaps.md (rationale + revisit trigger), never into the mechanical transform-rules.md (§16.10). Human-reviewed, never auto-merged.
What to report
The diagnosis, the options considered (including any rejected for violating a hard invariant), which was approved, what changed in which DR root, whether it has been re-planned, and the profile/region-gaps.md PR link.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/fix/SKILL.md
Popularity
- Repository stars
- 1
- 7-day trend score
- 3
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/mortennordbye/agentic-dr/fix)