agentleFS
hushh-labshushh-labs/hushh-research/.agents/skills/pkm-upgrade-rehearsal/SKILL.md

pkm-upgrade-rehearsal

Skill40 starsChanged 5 days agoApache-2.0View on GitHub

Use when proving zero-loss PKM upgrades with historical fixtures, reviewer rehearsals, rollback, scopes, and exact encrypted/decrypted artifacts.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    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:

    Avoid overlap with security-audit, reviewer-app-testing, and data-model-audit.

    line 12
  2. Distinctive:

    Treat structure-agent output as authorization to drop or overwrite information.

    line 45
  3. Distinctive:

    Run deterministic historical fixtures before intelligence, browser, or environment-backed checks.

    line 59
  4. Distinctive:

    Run the read-only active-shape audit and gated structure-agent chain without plaintext PKM in prompts.

    line 61
  5. Distinctive:

    Verify target-environment schema and route parity before requesting reviewer mutation authority.

    line 62
  6. Distinctive:

    When authorized, import the shared reviewer harness and run the app sequence without persisting keys or tokens.

    line 63

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Hussh PKM Upgrade Rehearsal Skill

Purpose and Trigger

  • Primary scope: pkm-upgrade-rehearsal
  • Trigger on PKM version/projection upgrades, preservation proof, reviewer payload rehearsal, quarantine, rollback, or scope-origin acceptance.
  • Avoid overlap with security-audit, reviewer-app-testing, and data-model-audit.

Coverage and Ownership

  • Role: spoke
  • Owner family: security-audit

Owned repo surfaces:

  1. .agents/skills/pkm-upgrade-rehearsal
  2. .agents/skills/pkm-upgrade-rehearsal/scripts/reviewer-pkm-app-rehearsal.mjs
  3. consent-protocol/scripts/audit_active_pkm_shape_readonly.py
  4. consent-protocol/scripts/eval_pkm_structure_agent.py
  5. scripts/ci/pkm-upgrade-gate.sh

Non-owned surfaces:

  1. consent-protocol/hushh_mcp/vault
  2. hushh-webapp/scripts/testing/reviewer-test-identity.mjs
  3. consent-protocol/db/migrations

Do Use

  1. Historical decrypt-transform-prove-encrypt-decrypt-rollback corpora.
  2. Occurrence-level preservation, equal-value deduplication, quarantine, rollback, and idempotency acceptance.
  3. Reviewer brokerage plus natural confirmed-memory rehearsal and exact payload artifacts.
  4. Canonical-scope stability and additive reserved/dynamic scope-origin metadata proof.

Do Not Use

  1. Generic reviewer login, BYOK, or navigation helpers; use reviewer-app-testing.
  2. Vault encryption or ordinary PKM storage implementation; use security-audit.
  3. Apply migrations, deploy, reset an account, or mutate shared fixtures without separate authority.
  4. Treat structure-agent output as authorization to drop or overwrite information.

Read First

  1. .agents/skills/reviewer-app-testing/references/byok-reviewer-browser-contract.md
  2. .agents/skills/pkm-upgrade-rehearsal/references/reviewer-pkm-upgrade-contract.md
  3. docs/architecture/pkm-cutover-runbook.md
  4. docs/architecture/pkm-storage-adr.md
  5. consent-protocol/docs/reference/personal-knowledge-model.md
  6. scripts/ci/pkm-upgrade-gate.sh

Workflow

  1. Classify the target contract/version and inventory every supported stored shape and active consumer.
  2. Run deterministic historical fixtures before intelligence, browser, or environment-backed checks.
  3. Require occurrence lineage, type/value preservation, quarantine restoration, rollback equality, and idempotency.
  4. Run the read-only active-shape audit and gated structure-agent chain without plaintext PKM in prompts.
  5. Verify target-environment schema and route parity before requesting reviewer mutation authority.
  6. When authorized, import the shared reviewer harness and run the app sequence without persisting keys or tokens.
  7. Accept only zero preservation, scope, authorization, coherence, export, and rollback failures.

Handoff Rules

  1. BYOK and browser-session mechanics route to reviewer-app-testing.
  2. Vault/storage implementation routes to security-audit.
  3. Migration governance routes to data-model-audit; deploy parity routes to uat-scoped-deploy.
  4. Broad security work returns to security-audit.

Required Checks

./scripts/ci/pkm-upgrade-gate.sh
./bin/hushh agents data-model-audit
./scripts/ci/reviewer-app-testing-check.sh

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.