wasikarn
@wasikarn
Public Skills
learn
by wasikarn
"Mines a session transcript for corrections, conventions and repeated workflows auto-memory missed. Use when wrapping up a session; not for saving one known fact."
ideate
by wasikarn
"Parallel divergent ideation (5 isolated agents, rotating frames, novelty/viability/fit scoring). Use when the question is open-ended. Say 'brainstorm'. Not for syntax, lookups, or closed-phrasing asks."
ste-lint
by wasikarn
"Checks prose against ASD-STE100 writing rules: sentence length, semicolons, contractions. Use when checking plain English or STE compliance. Don't use for AI-tell cleanup, see tech-humanize."
deep-audit
by wasikarn
"Deep-audit: post-implementation audit: verify every claim, score before/after, fix gaps, re-score. Use after an implementation pass. Not for plan conformance (/mh:compliance-audit) or first-pass review (/code-review)."
gate-report
by wasikarn
"Gate-report: ask/deny counts per gate from the gate-decisions journal, and mh skills unused in the last N sessions. Use for gate friction or stale skills."
harness-audit
by wasikarn
"Deterministic structural audit of the plugin's agents, skills, and hooks. Use when asked to audit the harness or check drift. Don't use for repo lint."
cost-report
by wasikarn
"Reports spend history per day, model and agent type from the cost-tracker log. Use when asked what sessions or subagents cost; not for live /usage."
model-bench
by wasikarn
"Compares two models' scored eval performance on the same case set via claude plugin eval --model, reusing existing graders. Use when choosing between models."
memory-lint
by wasikarn
"Lints the auto-memory store: dangling [[links]], orphans, index drift; --auto-archive trims. Use after editing memories or near the cap. Not for semantic or harness review."
tech-humanize
by wasikarn
"Humanize dev/tech writing (English/Thai) to sound natural, not AI-generated. Use when editing chat, standup/PR/commit, UI copy, or prose/ticket/spec/ADR. Don't use for translation or STE."
compliance-audit
by wasikarn
"Compliance-audit: verify a finished implementation against its plan; a fresh-context verifier reruns the gauntlet. Not for claim audits (/mh:deep-audit)."
idea-audit
by wasikarn
"Two isolated analysts plus a different-model attacker check an external source against evidence, then ship a scored adoption decision. Use after finding something worth adopting."
post-mortem
by wasikarn
"Writes the canonical record of a fixed bug. Use when the user says 'write a post-mortem' after a diagnosis. Don't use for an in-progress incident."
meta
by wasikarn
Use when testing check 78.
flat
by wasikarn
Use when testing check 75.
flat
by wasikarn
Use when testing check 78.
fixture-short-desc
by wasikarn
"Handles fixture provisioning workflows for check twenty regression testing. Use when validating the length gate stays silent under the 1536-char cap."
fixture-shared-name
by wasikarn
"Handles data export to CSV format for reporting teams needing scheduled batch downloads."
fixture-long-desc
by wasikarn
"Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only. Handles fixture provisioning workflows for check twenty length regression testing purposes only."
fixture-widget-tool
by wasikarn
"Provisions internal widget environments end-to-end for engineering teams who need a fast repeatable setup without manual configuration steps."
fixture-no-trigger
by wasikarn
"Handles bulk provisioning of widget environments for internal teams needing a quick repeatable setup path."
fixture-bad-yaml
by wasikarn
"Fixture for check 28 — deliberately malformed YAML frontmatter."
fixture-with-trigger
by wasikarn
"Handles bulk provisioning of widget environments. Use when a team requests a new environment without existing config."
fixture-boundary-20
by wasikarn
"12345678901234567890"