Mira diagnose verify

Skill eidos11/mira-adultedge/skills/mira-diagnose-verify

MIRA AdultEdge — Memory-Imagination-Reason Architecture for adult learning diagnosis and coaching

Install

npx -y skills add eidos11/mira-adultedge --skill mira-diagnose-verify

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

diagnosis + cross-model verification of reasoning text

SKILL.md

3.0 KB, 789 tokens by cl100k_base, as published. Nobody here has run it

mira-diagnose-verify

Run stages 1-3 of the MIRA pipeline (diagnosis + System B verification) on reasoning text and emit a structured diagnosis JSON — without coaching.

The coaching stage is intentionally absent: this skill produces the pre-coaching overlay so a caller can inspect, store, or pipe the diagnosis into mira-coach (skill ④) to obtain the full coaching response.

When to use

Trigger this skill when the user wants to:

diagnose reasoning text for cognitive patterns (PSR decomposition, pattern matching, System B overlay) without generating coaching output yet, or
pipe the diagnosis JSON into mira-coach for the full coached response.

Do not trigger for users who want a complete report in one step — run python -m mira --format json "<text>" followed by mira-report for that.

Output contract

The diagnosis JSON contains:

key	description
`schema_version`	contract version (mirrors `mira.report.SCHEMA_VERSION`)
`input`	the original learner text
`route_vtype`	verification type (`D` / `I` / `A`)
`psr`	`{P, S, R}` PSR decomposition
`patterns`	list of `{pattern_id, lane2_status, evidence_trace}`
`overlay_status`	overall verdict: `verified` / `structural_mismatch` / `insufficient_evidence`
`evidence_summary`	aggregated evidence text

Coaching is absent. Keys coaching and report are intentionally omitted.

2-step CLI usage contract

# Step 1 — diagnose reasoning text, emit pre-coaching JSON
python skills/mira-diagnose-verify/scripts/diagnose.py \
    "I failed because I'm not smart enough." \
    > diag.json

# Step 2 — pipe into mira-coach (skill ④) to obtain coaching
python skills/mira-coach/scripts/coach.py < diag.json > coached.json

Or via stdin in step 1:

echo "I failed because I'm not smart enough." | \
    python skills/mira-diagnose-verify/scripts/diagnose.py --stdin \
    > diag.json

Language flag:

python skills/mira-diagnose-verify/scripts/diagnose.py \
    --lang ko "시험에 떨어진 것은 내가 머리가 나빠서다." \
    > diag_ko.json

Structure

scripts/diagnose.py — thin wrapper: calls mira.pipeline.diagnose_only, serializes the diagnosis-subset JSON (no report, no coaching) to stdout.

Design notes

Thin wrapper only: all diagnosis logic lives in mira.pipeline.diagnose_only (stages 1-3). This script delegates entirely; it contains no engine logic.
Coaching absent by design: DiagnoseResult has no report or coaching field. The serializer (_build_diagnose_json) explicitly omits them.
Pattern: mirrors mira-report skill's script conventions — argparse, stdin-or-positional input, JSON stdout, error to stderr + exit 1.
Schema version: reuses mira.report.SCHEMA_VERSION ("0.2.2") so downstream tools (mira-report, mira-coach) can gate on the same version.

What ships with it: 1 file

5.2 KB alongside SKILL.md, 1 of them executable

scripts/

diagnose.pyruns5.2 KB