agentleFS
ashley-eyekyamashley-eyekyam/probematrix-learnings-coach/skills/supervisor-coach/SKILL.md

supervisor-coach

Skill0 starsChanged 27 days agoMITView on GitHub

Produce learning for first-line supervisors from a ProbeMatrix investigation export - shift in-charges, maintenance and operations supervisors, permit issuers, calibration signatories, foremen - covering what to check before work starts, what to verify before equipment returns to service, what to escalate and what to refuse, as checklists and a coaching guide. Use when someone asks what supervisors or shift leads should learn, for supervisor briefings, hold points, sign-off checks, or how to strengthen supervision after an incident.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Write the pack <run>/outputs/supervisor-coach.json (schema: learnings-coach/references/pack-schema.md), then:

    line 35

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Supervisor Coach

Audience: T2 first-line supervision. They decide which work starts and which work is accepted. In most ProbeMatrix analyses the Supervision level (L3) is the largest; this is where much of the learning lives.

Engine: PMLC = python3 <skills-root>/learnings-coach/scripts/pmlc.py (this skill needs the learnings-coach skill installed alongside it). Read learnings-coach/references/learning-principles.md.

Gather

PMLC select <ws> --tier T2 --brief
PMLC select <ws> --level 3 --top-rated                  # supervision factors that mattered most
PMLC select <ws> --tier T2 --relation approver                 # sign-offs and acceptances

Group L3 findings by sub-category: inadequate supervision, planned inappropriate operations, failure to correct a known problem, supervisory violations. These map to four supervisor habits: oversee, plan, correct, make the safe way the easy way. The last habit is about the system around the crew (tools, parts, time, clear hold points), not about policing people.

Write

  1. Bottom line: the five supervisor learnings from this event.
  2. Before work starts: a checklist of six to ten checks drawn from the findings (spec of parts, test requirements, permit scope, competence of the crew, known history of the equipment). Each check cites its evidence.
  3. Before equipment returns to service: hold points and acceptance criteria.
  4. Known problems: how to treat repeats (a third repair on the same item is a signal, not a routine) and when to escalate. Include who to escalate to.
  5. In an emergency: decisions a supervisor should be authorised and prepared to take without waiting, if the findings show delay from escalation. Present them as questions for the site to settle, not as authority the supervisor already has.
  6. What to refuse: situations from this event where a supervisor should decline to start or accept work, and the support they should expect from management.
  7. Recommendations and how supervisors will know they are working: one row per recommendation: what should change, the barrier, the hierarchy-of-controls tier, and a leading indicator a supervisor can read from their own shift records (e.g. "% of pressure-boundary jobs released with a design-pressure test record", "repeat repairs on the same tag escalated within 7 days", "notifications closed with materials and test data entered"). Every recommendation needs one; a recommendation without an indicator is not finished.
  8. Coaching conversations: three short scenarios from the event for supervisors to discuss in a team meeting, with prompts.

Output

Write the pack <run>/outputs/supervisor-coach.json (schema: learnings-coach/references/pack-schema.md), then:

PMLC check-pack <ws> <run>/outputs/supervisor-coach.json
PMLC pack <ws> <run>/outputs/supervisor-coach.json      # -> supervisor-coach.html (primary) and .md

Sections → kinds:

SectionKind
1 Bottom linebottom_line
2 Before work startscheck
3 Before return to servicecheck with "hold_point": true
4 Known problemspara
5 In an emergencyquestion, to the site
6 What to refusecheck
7 Recommendationsrecommendation, each with its indicator
8 Coachingscenario ×3

Set "scope": {"tiers": ["T2"]}. Give each check a group (Oversee, Plan, Correct, Make the safe way the easy way). Use data_quality, exclusions and supporting_cites as in pack-schema.md. The page prints the checks as a checklist.

{"kind": "check", "text": "Leak-test record attached and signed.", "hold_point": true,
 "when": "before return to service", "cites": ["FND-…"]}

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.