log-outreach
Skill1 starsChanged 2 months agoMITView on GitHub
Close out a session — reconcile both trackers against the files on disk, update PROJECT_NOTES with the phase and next action, and push anything vault-worthy to the Hermes vault. Run at the end of every working session.
In this repository37 other files it gives its agents.
- wsb-data-pipelineSkill
- docx-exportSkill
- harvard-mdxSkill
- humanise-docx-reportSkill
- make-pdfSkill
- bom-builderSkill
- ring-testSkill
- batch-draftSkill
- cold-emailSkill
- cv-intakeSkill
- cv-upgradeSkill
- find-contactSkill
- followupSkill
- humanize-emailSkill
- outreach-statusSkill
- target-researchSkill
- erc-statusSkill
- log-sessionSkill
- score-auditSkill
- submission-checkSkill
- sumo-statusSkill
- trial-runSkill
- vr-statusSkill
- academic-registerSkill
- aviation-data-sourcesSkill
- business-frameworksSkill
- consultancy-registerSkill
- excel-deliverablesSkill
- forecasting-methodsSkill
- operations-frameworksSkill
- secondary-data-sourcesSkill
- archive-indexSkill
- event-proposalSkill
- meeting-minutesSkill
- read-boxnoteSkill
- captureSkill
- new-scriptSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 5 are distinctive, and 3 are 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 16
Update statuses: Every draft has a status in its frontmatter that says exactly what still has to happen to it.
- Distinctive: line 27
Push the vault-worthy parts: Not everything belongs in the vault.
- Very specific: line 34
Read _CLAUDE.md at the vault root first.
- Very specific: line 36
Mirror to Codex if the brain changed: If anything in .claude/agents/ or .claude/rules/ was edited this session:
- Distinctive: line 70
Never log work that did not happen.
- Distinctive: line 71
Never mark a phase complete when its deliverable is not on disk.
- Distinctive: line 72
Never write a contact's email address or phone number into the vault.
- Distinctive: line 74
If the trackers and the files disagree repeatedly across sessions, say so in the report.
≈ “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.
Files
/log-outreach, close the session
Usage: /log-outreach
The next session reads PROJECT_NOTES.md first. Whatever this skill writes is what the next session believes, so it has to be true.
Steps
- Reconcile the trackers against disk. For each person, compare
tracker.mdagainst the files indrafts/andsent/. The files win. Fix the tracker where they disagree, and note what was out of step — a recurring mismatch means a step in the workflow is being skipped. - Update statuses. Every draft has a
statusin its frontmatter that says exactly what still has to happen to it. Never end a session with a file in an unclear state. - Set follow-up dates. Anything sent this session gets
followup_dueat +6 days, in the frontmatter and the tracker. - Update
PROJECT_NOTES.md:- The phase, and honestly — a phase whose deliverable does not exist is not done
- What is blocked, and who unblocks it
- What is still undecided
- One next action
- A new row in the session log: date, and what actually happened
Push the vault-worthy parts. Not everything belongs in the vault; this repo holds the operational detail. The vault gets:
- A dev-log note in
10 Logs/for a substantial session →/obsidian-log - A person note in
04 People/for anyone who replied →/obsidian-person. Name, company, role, context. No email address, no phone number - An interview or call → one line in
09 Reference/Planned Events.md - A decision about strategy or targeting →
/obsidian-decide - Anything else worth keeping →
/obsidian-save
Read
_CLAUDE.mdat the vault root first. Its folder map and frontmatter schemas are authoritative.- A dev-log note in
- Mirror to Codex if the brain changed. If anything in
.claude/agents/or.claude/rules/was edited this session:cp .claude/agents/*.md .codex/agents/ cp .claude/rules/*.md .codex/rules/ - Report what was written.
Output format
## Session logged — <date>
**Did this session**
- <what actually happened>
**Trackers reconciled**
- <Person Name>: <n> rows, <n> corrected
- <Second Person>: <n> rows, <n> corrected
- Mismatches found: <what, or none>
**PROJECT_NOTES**
- Phase: <n> (<changed from / unchanged>)
- Next action: <one thing>
**Vault**
- <note> → <path>
- Or: nothing vault-worthy this session
**Codex mirror:** synced | not neededRules
- Never log work that did not happen. An honest empty session is a session where the notes stay true.
- Never mark a phase complete when its deliverable is not on disk.
- Never write a contact's email address or phone number into the vault.
- One next action in PROJECT_NOTES.
- If the trackers and the files disagree repeatedly across sessions, say so in the report. That is a workflow bug, not a clerical one.
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
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.

