agentleFS
patrickbdevaneypatrickbdevaney/hermes-max/skills/workflow-quality-bar/SKILL.md

workflow-quality-bar

Skill1 starsChanged 6 months agoMITView on GitHub

Software implementations must meet senior-engineer review standards: complete type annotations, docstrings on public classes/methods, explicit error handling with informative messages, no placeholder comments or TODOs in committed code, and tests covering edge cases — not just the happy path. Applies to the planner (the plan must SPECIFY these) and the executor (the code must HAVE them). quality_check surfaces the gaps as advisory warnings; it never replaces the hard verify gate.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are 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.

  • 2

    Very specific

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

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

    no placeholder comments, no TODO/FIXME left in committed code.

    line 32
  2. Very specific:

    Run mcp_hermes_max_verify_quality_check(path) on a Python file you implemented.

    line 45

≈ “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
<!-- TRIGGERS WHEN: about to declare a Python implementation done, or writing the plan contract — raise the output to senior-review texture (advisory, not a gate) -->

Quality bar — senior-review texture, advisory

The deterministic gate (lint + types + tests, [[workflow-verify-enhanced]]) proves code is correct. It does not prove it is senior-grade. A finished implementation has no gap a senior reviewer would flag:

  • complete type annotations on public functions/methods;
  • docstrings on all public classes and methods;
  • explicit error handling with informative messages (no silent except:);
  • no placeholder comments, no TODO/FIXME left in committed code;
  • tests covering edge cases, not just the happy path.

Applies to both phases of the plan/execute split

  • Planner ([[workflow-plan-contract]]): the PLAN.md FILE SPECs must specify these — exact typed signatures, the error type/message per edge case, the edge-case test names. If the plan demands them, the executor produces them.
  • Executor ([[workflow-execute-from-plan]]): the code must have them. Before declaring a file done, run the advisory check below and close what it flags.

The advisory check (quality_check)

Run mcp_hermes_max_verify_quality_check(path) on a Python file you implemented. It flags, without failing the build:

  • annotations_missing — params/return with no type annotation;
  • docstrings_missing — public functions/methods with no docstring (dunders exempt);
  • placeholders — TODO/FIXME/placeholder/stub markers;
  • bare_excepts — except: with no type (swallows everything).

clean: true means none of the above. Treat each finding as a texture gap to close, not a verdict.

The line that must not blur

quality_check is advisory — it raises texture, it does not gate. The hard pass/fail stays verify() / deep_verify() ([[workflow-verify-enhanced]]) and the DONE_CONDITION ([[workflow-done-definition]]). Never block a green build on a style warning; never declare done while a senior reviewer would still flag the code.

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.