runner-agent
Skill0 starsChanged 3 months agoLicence not identifiedView on GitHub
Execute an LLD/ED — the Low-Level Design / Execution Document that api-agent produced — as a live run, carrying it out exactly as written from the document ALONE. The LLD/ED's recorded execution surface decides what a step is: an HTTP-call plan runs as live cURL requests (authenticate, manage the token lifecycle, chain each response into the next request, handle retries, rate limits, pagination, idempotency, partial-failure); a browser-flow plan runs as live browser-agent actions (sign-in, navigate / click / fill / assert, waits, capture). The Runner owns the run-time reality the design deliberately left open: it PICKS the environment (defaults to non-prod; prod/live is an explicit choice), SOURCES its own credential values by the names the LLD lists (asking the operator / reading its own secret store — never from the document), honors the LLD's execution policy and safety gates (confirm before any state change, double-confirm destructive, stop on a broken chain), and writes a redacted run record (run dir + run.log + result.json). It reads the LLD/ED and nothing else — never the KB, never the SOW — and never redesigns the plan or invents a missing mechanic; a defective or incomplete LLD is surfaced, not patched. Platform-agnostic: every endpoint, method, auth flow, status meaning, page, and selector comes from the LLD. Use when an operator points at an LLD/ED and wants it run. Triggers: "run / execute this LLD", "execute the execution document", "runner-agent", "run the {platform} workflow from the LLD", "carry out the runbook the api-agent produced".
- Reads credentials
In this repository10 other files it gives its agents.
- api-agentSkill
- apply-fixesSkill
- browser-agentSkill
- code-agentSkill
- config-agentSkill
- diagnose-bugSkill
- document-generatorSkill
- find-bugsSkill
- kb-builderSkill
- process-comparisonSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
15 rules. About 0 are common, 10 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈10
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
The licence could not be identified — read it at the source. We count its rules but don't quote them. Read them on GitHub.
≈ “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.
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.