brief
Session bootstrap — loads shared memories and team standards from the Mori server.
Skill by fjwood69
Chalk Labs
Need help deploying brief?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/fjwood69-brief
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: backend
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/brief
What's inside
- Skill name
brief- Description
- Session bootstrap — loads shared memories and team standards from the Mori server. Use at session start, or after a context compaction with --post-compact.
- License
- AGPL-3.0
- Instructions
- 99 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
**If the Mori server is unreachable** (a tool call returns a connection error, ECONNREFUSED, or timeout), do not show a stack trace. Tell the user the Mori server appears to be down and that a running Mori server is required. Point them to the quickstart at https://github.com/fjwood69/mori#quickstart and suggest checking the server URL in plugin settings. Then stop.
## Argument Parsing
Parse the raw input for:
- `--project <name>`: scope the brief to a specific project
- `--auto`: detect project from the current git working directory
- `--post-compact`: lightweight **delta** re-grounding after a context compaction —
surfaces only what changed in shared state since the last brief, not the full base.
Triggered automatically by the Mori `SessionStart` hook when `source == "compact"` —
Claude Code re-fires `SessionStart` after compaction with that source field, which is
the only supported mechanism for injecting context at that seam. Can also be run manually.
If no arguments: run the standard unscoped brief.
## Marker file
The brief boundary is tracked in a per-config-dir marker:
```
${CLAUDE_CONFIG_DIR:-$HOME/.claude}/.mori-last-brief
```
It holds the UTC ISO-8601 timestamp of the most recent brief. Stamp it (overwrite
with the current UTC time) at the **end** of every brief run, in all modes:
```bash
date -u +%Y-%m-%dT%H:%M:%SZ > "${CLAUDE_CONFIG_DIR:-$HOME/.claude}/.mori-last-brief"
```
## Execution
### 1. Resolve project (if --auto or --project)
**If `--auto`:**
1. Check for `.mori-project` file — walk up from CWD to filesystem root; if found, use its text content (stripped, lowercased)
2. Else check `MORI_PROJECT` environment variable
3. Else run `git rev-parse --show-toplevel 2>/dev/null` and use `basename` of the result (lowercased)
4. If no project detected: fall back to unscoped brief and note "no project detected, loading unscoped"
**If `--project <name>`:** use that name directly.
### 2. Call the MCP tool
**Standard brief (default):**
- With project: call `mori-brief` with `project=<name>`
- Without project: call `mori-brief` (no params — loads all memories up to cap)
**Post-compact brief (`--post-compact`):** resolve the `since` boundary first
(session-aware), then call the delta tool.
1. **Read the marker** `${CLAUDE_CONFIG_DIR:-$HOME/.claude}/.mori-last-brief`. If it
exists and is non-empty, use its contents as `since`.
2. **Else derive session start** (best-effort): the start time of the current session
transcript `.jsonl` (e.g. the file's first-event timestamp, or its creation time).
3. **Else omit `since`** and let the server default window (`MORI_POST_COMPACT_WINDOW`,
default `6h`) apply.
Then call `mori-brief` with `post_compact=true`, plus `project=<name>` if resolved and
`since=<resolved>` if found:
```
mori-brief(post_compact=true, project="<name>", since="<iso-or-omitted>")
```
The delta lists are capped; if the tool reports "…N more — run a full /brief if you
need the rest", relay that pointer rather than auto-running a full brief.
### 3. Check pending messages
Call `mori-msg_recv(unacked=True)` (both modes — new messages are part of the delta).
If messages are returned, surface them after the memory summary:
- **task / question** — prominent; include a ready-made reply command:
- task → `mori-msg_send(to="<from_host>", type="ack", reply_to="<id>", body="acknowledged")`
- question → `mori-msg_send(to="<from_host>", type="reply", reply_to="<id>", body="...")`
- **decision / broadcast** — awareness items only (lower prominence)
Skip this section silently if no pending messages or if `mori-msg_recv` fails (daemon may not be running).
On `--post-compact`, also catch cross-device traffic that arrived during the session:
call `mori-nats_sub(replay=true, wait=2)` and surface anything new.
### 4. Stamp the marker
Write the current UTC time to the marker file (see **Marker file** above). Do this in
both modes, after the tool calls.
### 5. Report
**Standard brief:** report "Ready" — summarise what was loaded (memory counts, project
scope, dream state, pending message count if any). Do not take autonomous actions.
**Post-compact brief:** report a one-line **abbreviated** re-grounding summary, e.g.
"Re-grounded — N changed, M superseded, K messages". Do not re-dump the full base; the
working context is already preserved by the compaction summary.From fjwood69/mori, AGPL-3.0.
More from fjwood69/mori
- 0.649score
ingest
fjwood69
Bootstraps the memory store from existing material (repos, PDFs, transcripts).
SkillAFreshness: A★ 32Claude Code +9 - 0.649score
consult
fjwood69
Strategic guidance from the advisor model on a question, with optional focus area and depth.
SkillAFreshness: A★ 32Claude Code +9 - 0.649score
export
fjwood69
Bundle the canonical memory set into one structured Markdown document for external-LLM review, audit, or dashboard download.
SkillAFreshness: A★ 32Claude Code +9 - 0.646score
wrap
fjwood69
Session wrap — summarise session, publish to cc-share and NATS, then dream
SkillAFreshness: A★ 32Claude Code +9 - 0.646score
msg
fjwood69
Inter-agent messaging — send tasks, questions, and decisions to other Mori agents over NATS
SkillAFreshness: A★ 32Claude Code +9 - 0.646score
req
fjwood69
Lightweight project requirements checklist surfaced via /brief.
SkillAFreshness: A★ 32Claude Code +9 - 0.644score
pensieve
fjwood69
Ranked full-text search and browse over the shared memory store.
SkillAFreshness: A★ 32Claude Code +9 - 0.644score
dream
fjwood69
Runs the dream pipeline that distils session events into durable memories.
SkillAFreshness: A★ 32Claude Code +9 - 0.644score
nats
fjwood69
Cross-device messaging over the NATS bus — ping, publish, subscribe.
SkillAFreshness: A★ 32Claude Code +9
Signals
- Stars
- 32
- Forks
- 5
- Open issues
- 14
- Last commit (this folder)
- Jun 15, 2026 (4 months ago)
- Last release
- Oct 3, 2026 (29 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- AGPL-3.0
- Repository
- github.com/fjwood69/mori
- Website
- moriapp.dev/
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
More from fjwood69
- 0.649score
ingest
fjwood69
Bootstraps the memory store from existing material (repos, PDFs, transcripts).
SkillAFreshness: A★ 32Claude Code +9 - 0.649score
consult
fjwood69
Strategic guidance from the advisor model on a question, with optional focus area and depth.
SkillAFreshness: A★ 32Claude Code +9 - 0.649score
export
fjwood69
Bundle the canonical memory set into one structured Markdown document for external-LLM review, audit, or dashboard download.
SkillAFreshness: A★ 32Claude Code +9 - 0.646score
wrap
fjwood69
Session wrap — summarise session, publish to cc-share and NATS, then dream
SkillAFreshness: A★ 32Claude Code +9 - 0.646score
msg
fjwood69
Inter-agent messaging — send tasks, questions, and decisions to other Mori agents over NATS
SkillAFreshness: A★ 32Claude Code +9 - 0.646score
req
fjwood69
Lightweight project requirements checklist surfaced via /brief.
SkillAFreshness: A★ 32Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get brief running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
brief skill. Agent Mart. https://agentmart.dev/skills/fjwood69-brief (last updated 2026-10-24)