target-research
Skill1 starsChanged 2 months agoMITView on GitHub
Research a target company and write a dossier into shared/companies/ — what they actually build, which team an intern would join, recent specific news, and the overlap with the candidate. Use before writing any first-contact email, when deciding whether a company is worth targeting, or when preparing for an interview.
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
- log-outreachSkill
- outreach-statusSkill
- 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
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
Check for existing work: Read shared/companies/<slug>.md if it exists.
- Very specific: line 16
Add the company to
shared/targets.md: if it is not there. - Distinctive: line 66
Never invent a recent event.
- Distinctive: line 68
Do not research companies not on the target list.
≈ “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
/target-research, build the company dossier
Usage: /target-research Space42 · /target-research "Technology Innovation Institute" --for "<Person Name>"
The dossier is what makes paragraph 1 of the email possible. Without it, there is only a template with a name pasted in.
Steps
- Check for existing work. Read
shared/companies/<slug>.mdif it exists. If it was researched inside the last month, refresh the recent-news section rather than starting over. Also run/obsidian-find <company>— the vault may already know something. - Add the company to
shared/targets.mdif it is not there. - Read the candidate's
evidence.mdbefore searching. The point is to find overlap, and it is much easier to spot overlap when the search already knows what to look for. - Delegate the search to the
company_researcheragent. It has the full method. It reads job postings before marketing pages, because job postings are where the real stack is written down. - Look specifically for:
- Open engineering roles, including senior ones. They list the stack
- The engineering blog, GitHub org, published papers
- Whether they run an internship programme and through what route
- One recent, dated, specific, linkable event
- Which team the candidate would plausibly sit in
- Write the dossier to
shared/companies/<slug>.mdin the format specified inagents/company_researcher.md. - Assign a tier and record why:
- Tier 1 — strong technical overlap, plausible route in, worth a properly researched email
- Tier 2 — real overlap but a harder route, or a less exact fit
- Tier 3 — worth an application through their formal process, not worth a cold email
- Say if the answer is no. A company with no overlap or no reachable route is a legitimate finding. Recording it stops the next session repeating the search.
Output format
## Researched: <company> — Tier <n>
**What they build:** <two sentences of engineering reality>
**Stack:** <from job postings>
**Team to target:** <name>
**Takes interns:** yes | no | formal programme only | unknown
**Paragraph-1 material**
- <specific recent fact> — <URL> (as of YYYY-MM)
**Overlap with <person>**
| They need | <person> has | Strength |
|---|---|---|
**Angle:** <one sentence>
**Gaps:** <what could not be verified>
**Recommendation:** cold email | formal application | skip
**Next:** /find-contact <company>Rules
- Job postings before marketing pages. Always.
- Date every unstable fact. Team structures and funding move quarterly.
- Never invent a recent event. "No recent public activity found" is a real answer and it changes the angle rather than blocking it.
- Mark inferences as inferences.
- Do not research companies not on the target list. Add them there first so the list stays honest.
- One dossier per company, shared by both people. The overlap section gets a subsection per person.
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.

