phrazzld/orient
skills/orient/SKILL.md
Fast session-start repository orientation from live local evidence. Use when: "orient yourself", "start of session", "new session", "where are we", "catch me up before acting", after compaction, after switching worktrees, or before choosing a Harness Kit workflow. Trigger: /orient, /ground, /session-start.
$ install --global
skillsauthnpx skillsauth add phrazzld/agent-skills orientInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Jul 2, 2026, 6:14 AM142.7s2 files scanned
SKILL.md
- name:
- orient
- description:
- |
- argument-hint:
- [scope|--deep]
/orient
Ground the session in repo truth — not memory, not a stale summary. Read-only, fast.
What it produces
A short, natural read of where things stand and the one move that follows. You are writing to stop a wrong first step, not to fill a form. In plain prose, cover only what the evidence supports:
- where we are (repo, branch, clean vs dirty) and what's in flight;
- the single best next move, why it wins, and what would make it wrong.
Punchy beats complete. A sentence or two when the state is obvious; a short list
only when it's genuinely tangled. No fixed template, no field-by-field rundown,
no history dump. If the user already gave a precise command and the state is
clear, that sentence is the whole report. --deep is the only license to dig.
Example, obvious state: "On master, clean, CI green. Shaped ticket 058 is the
only thing in flight and nothing blocks it — next is /deliver 058."
Where the signal lives
Read the smallest set that explains the workspace; skip whatever the prompt
already settled. Beyond the obvious (git status, branch, recent commits):
- scoped then repo
AGENTS.md;project.mdfor focus (README.mdonly if still unclear) backlog.d/*.mdfor active work;backlog.d/_done/*andCloses-backlog:trailers for closure.harness-kit/agents.yaml/ roster probe — only when the next move is a delegation.harness-kit/agent-readiness.yaml— only to name the profile state, never to score the repo
Routing
Recommend one next skill. If the signal is mixed, name the missing evidence rather than guess from vibes.
| Signal | Next |
|---|---|
| Shaped ticket, clean branch | /deliver <ticket> |
| Dirty branch, intended changes | /deliver --polish-only |
| Empty or stale backlog | /groom |
| Failure, broken gate, unclear root cause | /diagnose |
| Finished work needing closeout | /ship, /yeet, or /reflect |
| Readiness or profile question | /qa or the repo's readiness surface |
| "What happened / why does it matter" | /orient --deep or /shape |
| Skill or harness primitive change | /harness-engineering |
Stay in lane
Orient reads; it never acts. Don't deliver, groom, refactor, reflect, debrief,
mine transcripts, or mutate state. Don't label the repo "ready" or "validated" —
route that to the owning skill. Don't store session memory; durable state lives
in backlog, commits, and receipts. Don't spin up provider lanes unless scope is
broad, stale, or contested. Scoped AGENTS.md governs — respect it.
Verification
Editing this skill in the Harness Kit source repo:
cargo run --locked -p harness-kit-checks -- check-frontmatter --repo .
Invoked in another repo, that command isn't required — acceptance is the report's cited live evidence and a next move the operator can act on.
Related Skills
phrazzld/compound
testing
VerifiedTrustedCommunity
Capture one compounding repo-technical learning while a solved problem is still fresh. Use when: after a bug fix, diagnosis, delivery, review, or incident reveals a reusable pattern worth adding to `docs/solutions/`. Trigger: /compound, /capture-learning, /learning.
13SKILL.mdUpdated Jul 5, 2026
phrazzld/factory-apps
testing
VerifiedTrustedCommunity
Route Misty Step factory application capabilities. Use when choosing, auditing, integrating, or operating Canary, Powder, Landmark, Aesthetic, or Bitterblossom: production observability, incidents, health checks, error logging, backlog/work-card state, release intelligence, UI/UX system adoption, or supervised/unsupervised agent dispatch. Trigger: /factory-apps, /factory-stack.
13SKILL.mdUpdated Jul 4, 2026
phrazzld/skill-eval
testing
VerifiedTrustedCommunity
Prove a skill beats no-skill with a falsifiable A/B eval, or retire it. Design, generate, run, and maintain a skill-specific eval: name the one claim the skill must earn, run it skill-on vs raw same-model, grade blind with objective checks first, return a keep/adapt/cut verdict. Use when: "eval this skill", "does this skill help", "prove the skill beats no skill", "write an eval for", "benchmark a skill", "is this skill worth it", "skill A/B", "skill regression test", "generate skill evals". Trigger: /skill-eval, /eval-skill, /prove-skill.
13SKILL.mdUpdated Jul 2, 2026
phrazzld/skills/harness-engineering/templates/repo-local-skill
tools
VerifiedTrustedCommunity
> Template. Copy to `<target-repo>/.agents/skills/<repo>-<domain>/SKILL.md` > and fill every bracketed placeholder from the live target repo. Delete this > line and every other `> ` guidance line before committing. See > `../../references/repo-local-skill-generation.md` for the full process. --- name: <repo>-<domain> description: | [One paragraph: what this skill verifies/runs/operates for <repo>, stated in terms of the repo's real shape (service/CLI/library/etc.), not generic process. En
13SKILL.mdUpdated Jul 2, 2026