skills/claude-code-audit-workspace/iteration-1/eval-3-specific-leverage-framing/with_skill/outputs/drafts/skills/reorient/SKILL.md
Produce a 5-line situational summary on demand or whenever the user is resuming a session, catching up, or lost. Triggers on "resume", "where are we", "what were we doing", "status", "remind me", "catch me up", or immediately on SessionStart:resume. Replaces Claude's default behavior of dumping the entire saved continuity record back at the user — instead, always outputs exactly 5 lines in a fixed shape so the user can reorient in under a second. Pairs with circuit:handoff continuity records when present but does not require them.
npx skillsauth add petekp/claude-skills reorientInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Any of:
resume, resume handoff, continue, continue from handoff, or /circuit:handoff resume (with or without args).Output exactly 5 lines. In this exact shape:
Goal: <one sentence, the highest-level thing we are trying to accomplish this session or arc>
Last: <one sentence, the last meaningful action taken — what just finished or was just attempted>
State: <one sentence, the current state of the work — what exists on disk, what's green, what's red>
Next: <one sentence, the single next action>
Open: <one sentence OR "none" — blockers, decisions awaiting your input, or unresolved ambiguity>
No preamble. No trailing summary. No headers. No markdown other than backticks around file paths or commit SHAs. If the answer to any line is genuinely unknown, write Goal: unknown — reading session context to reconstruct (or the equivalent for that line) and then immediately investigate. Never output speculation as fact.
.circuit/control-plane/continuity-records/ has a pending record, read its narrative.goal, narrative.next, and git.head fields. These are authoritative for Goal / Next / part of State.TaskList if tasks exist — the most recently updated in_progress task IS the Next line.git log -1 --format=%h\ %s gives the last commit (part of State). git status --short (part of State).Do NOT read the full saved continuity payload and dump it. Read the narrative fields, extract 5 facts, stop.
unknown on that line and go find out after.Open: line.Input: User opens a fresh session, first message is resume.
Saved continuity record's narrative says goal = "Close ADR-005 Phase 3 by migrating runtime_stats.rs onto observation::transcript." TaskList has task #3 (in_progress) "Wire ingest_transcript_observation into runtime_stats counter." Git HEAD is b7c0b0b9 phase-3-q4 landed. git status is clean.
Output:
Goal: Close ADR-005 Phase 3 by migrating runtime_stats.rs onto observation::transcript.
Last: Landed Phase 3 Q4 (b7c0b0b9); working tree clean.
State: Migration not started — runtime_stats.rs still reads transcripts directly; #[allow(dead_code)] tripwire on ingest_transcript_observation still in place.
Next: Wire ingest_transcript_observation into runtime_stats counter (TaskList #3 in_progress).
Open: none — signed-off to proceed.
Then stop. Wait for the user's next message.
If circuit:handoff exists and has a pending record, reorient reads it but does not re-save or advance continuity state. If the user types resume handoff or /circuit:handoff resume, invoke circuit:handoff for its bookkeeping side effects — but the user-facing output is still the 5-line reorient block, not the handoff skill's default dump.
In effect: reorient is the presentation layer; circuit:handoff is the persistence layer. One is not the other.
tools
Comprehensively manually test the Circuit plugin's user-facing surface in either Claude Code or Codex. Use this skill whenever the user asks to "manually test Circuit", "QA the Circuit plugin", "exercise the Circuit surface", "run the Circuit checklist", "smoke test Circuit", "find regressions in Circuit", "test the Claude Circuit plugin", "test the Codex Circuit plugin", or when preparing a Circuit release for marketplace publication. Argument is the host package to test — `claude` or `codex`. Produces a Markdown report with per-command pass/fail, exploratory findings ranked by severity, run-folder evidence links, and a concise terminal summary. Use even if the user does not say the word "test" — phrases like "go through every Circuit command" or "make sure Circuit still works end-to-end" should also trigger.
development
Turn the prompt supplied with this skill into a concise, auditable Codex Goal or explain why a Goal is not the right fit. Use when the user asks to draft, formulate, rewrite, tighten, or create a `/goal` from a plain-language task, especially for multi-step work that needs a durable objective, evidence-based completion, constraints, iteration policy, and a default adversarial review loop.
development
Give the human a fast, plain-English catch-up on what changed in the project: what the agents did, why, and what decisions need their input. Use this whenever the user asks to "catch me up", "what changed", "where are we", "recap", "brief me", "give me the rundown", "what did you do", "summarize the session", "fill me in", or otherwise signals they have been away and want to get back up to speed quickly. Built for someone steering several agent-driven projects at once who does not read the code closely but needs to grasp the core ideas, the choices made, and the open decisions well enough to steer. Trigger even if they do not use these exact words: any request to get oriented on recent progress should use this skill.
tools
Expert Unix and macOS systems engineer for shell scripting, system administration, command-line tools, launchd, Homebrew, networking, and low-level system tasks. Use when the user asks about Unix commands, shell scripts, macOS system configuration, process management, or troubleshooting system issues.