autonomous-unblock
Skill4 starsChanged 8 months agoMITView on GitHub
Keep progress moving when work stalls on blockers. Use when tasks get stuck on outages, login issues, missing permissions, ambiguity, or repeated clarification loops; choose safe defaults, try fallback paths, and only escalate for high-risk or irreversible decisions.
In this repository50 other files it gives its agents.
- model-usage/CLAUDE.mdCLAUDE.md
- peekaboo/CLAUDE.mdCLAUDE.md
- apple-notesSkill
- apple-remindersSkill
- bear-notesSkill
- birdSkill
- blogwatcherSkill
- blucliSkill
- camsnapSkill
- coolifySkill
- easydmarcSkill
- eightctlSkill
- email-deliverySkill
- food-orderSkill
- geminiSkill
- gifgrepSkill
- githubSkill
- gogSkill
- goplacesSkill
- himalayaSkill
- imsgSkill
- local-placesSkill
- mcporterSkill
- model-usageSkill
- namecheapSkill
- nano-pdfSkill
- notebooklm-researchSkill
- notionSkill
- obsidianSkill
- openai-image-genSkill
- openai-whisperSkill
- openhueSkill
- oracleSkill
- ordercliSkill
- peekabooSkill
- pieces-mcpSkill
- railwaySkill
- sagSkill
- send-payloadSkill
- skill-creatorSkill
- slack-signal-monitorSkill
- songseeSkill
- sonoscliSkill
- spotify-playerSkill
- summarizeSkill
- twilioSkill
- vercelSkill
- video-framesSkill
- vip-emailSkill
- weatherSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Keep working when blocked instead of waiting for user direction.
- Distinctive: line 17
If low-risk details are missing, pick a reasonable default assumption, state it briefly, and proceed.
- Distinctive: line 22
Ask the user only for:
- Distinctive: line 29
When escalation is required, ask one specific unblock question that includes:
≈ “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
Autonomous Unblock
Keep working when blocked instead of waiting for user direction.
Execution Protocol
- Restate the objective in one sentence.
- Attempt recovery paths in this order:
- retry with corrected parameters or narrower scope
- switch to an alternate tool, provider, or platform
- generate a momentum-preserving fallback artifact (draft, checklist, spec) and continue
- If low-risk details are missing, pick a reasonable default assumption, state it briefly, and proceed.
- Escalate only when required by risk boundaries.
Escalation Boundaries
Ask the user only for:
- irreversible or destructive actions
- spending money or binding external commitments
- legal, compliance, privacy, or safety risk
- mandatory human verification (captcha, 2FA, account ownership checks)
When escalation is required, ask one specific unblock question that includes:
- what is blocked
- what has already been attempted
- the single action required from the user
Then continue any remaining unblocked work in parallel.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.