Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/spacedock-dev/spacedock/debriefnpx skills add spacedock-dev/spacedock --skill debriefgit clone --depth 1 https://github.com/spacedock-dev/spacedockWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00059 | $0.05295 |
| Opus 5 | $0.00030 | $0.02648 |
| Sonnet 5 | $0.00012 | $0.01059 |
| Haiku 4.5 | $0.00006 | $0.00530 |
Grade A, and why
debrief scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 414 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Session Debrief
You are producing a session debrief — a structured record of what happened during a workflow session. The debrief captures commits, task state changes, gate decisions, issues found, and observations. It feeds forward into the next session so the first officer starts with context instead of cold.
Follow these phases in order. Do not skip or combine phases.
Phase 1: Discovery
Launcher command invariant: Prefer ${SPACEDOCK_BIN:-spacedock} for Spacedock helper calls so an explicit front-door launcher remains the command source; when unset, fall back to spacedock on $PATH. Bare spacedock examples are shorthand for that expression.
Step 1 — Identify the workflow
The user may provide a workflow directory path as an argument. If they didn't, search for it:
- Run
${SPACEDOCK_BIN:-spacedock} status --discover. It prints one resolved workflow directory path per line on stdout, scanning from the enclosing git repository root and pruning agent-managed worktrees (.claude/worktrees/) and other build/vendor noise. - If exactly one path is returned, use it as
{dir}. If multiple, ask which one. If none, report "No Spacedock workflow found."
Store the confirmed path as {dir}.
Step 1b — Resolve the debrief home
The debrief home is not always {dir}. For a split-root workflow (its README declares a state: checkout) the established _debriefs/ history and the auto-syncing convention live in the state checkout, not the definition dir on the code branch. Let the binary tell you which it is rather than re-parsing state: yourself:
- Run
${SPACEDOCK_BIN:-spacedock} status --boot --json --workflow-dir {dir}once and read three fields the binary already resolves:state_backend(single-root|split-root),entity_dir(the absolute resolved debrief home), andentity_dir_present. - Set
{debrief_root} = entity_dir. Use{debrief_root}/_debriefs/everywhere_debriefs/appears below.
state_backend == single-root→entity_diris{dir}; debriefs read, write, and commit under{debrief_root}/_debriefs/on the current branch (unchanged behavior).state_backend == split-root→ set{state_checkout} = entity_dir(= {debrief_root}). Halt-gate: ifentity_dir_presentisfalsethe declared checkout is NOT materialized (an orphan state branch with no linked worktree — a fresh clone or a removed worktree). HALT, report "state not initialized," and stop. Do NOT fall back to writing into{dir}on the code branch. Tell the captain to runspacedock state init(manual fallback:git fetch origin {state_branch} && git worktree add {state_checkout} {state_branch}), then re-run the debrief. Otherwise the reads, the new-file write, and the commit all run in the state checkout on its own branch — never in{dir}on the code branch. Resolve the state branch from the checkout itself:git -C {state_checkout} rev-parse --abbrev-ref HEAD.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 414 lines · 59 tokens per session scan A 99342434a141
debrief is a skill published in the GitHub repository spacedock-dev/spacedock (75 stars, last pushed 3d ago), licensed Apache-2.0. It adds 59 tokens to every session and 5,295 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
fleet
Parallel campaign orchestrator. Runs multiple campaigns in coordinated waves within a single session. Spawns 2-3 agents per wave in isolated worktrees, collects discoveries, shares context between waves. Use when work decomposes into 3+ independent streams that can run simultaneously.
archon
Autonomous multi-session campaign agent. Decomposes large work into phases, delegates to sub-agents, reviews output, and maintains campaign state across context windows. Use for work that spans multiple sessions and needs persistent state, quality judgment, and strategic decomposition.
setup
First-run experience for the harness. Three modes: Recommended (guided, 3 min), Full Tour (guided + skill walkthrough, 8 min), and Express (zero questions, 30 sec). Installs hooks first, detects stack, configures harness.json, runs a live demo on real code, and prints a reference card.
daemon
Continuous autonomous operation mode. Keeps campaigns running 24/7 by chaining Claude Code sessions via RemoteTrigger. Each session picks up from the campaign's continuation state, works until context runs low or the phase completes, then schedules the next session. Auto-stops on campaign completion or budget…
dashboard
Real-time harness observability dashboard. Reads campaigns, fleet sessions, telemetry, and pending queues to present a snapshot of harness state at a glance. Invoked by /dashboard, /do status, or phrases like "what's happening" and "show activity".
evolve
Research-driven multi-cycle improvement director. Forms causal hypotheses about why scores are low, validates them with scout agents before attacking, dispatches axis-parallel fleet attacks, extracts transferable patterns, and runs indefinitely within a budget envelope. Accumulates a persistent belief model and…