OpenTSC
Local CLI vault for contact intelligence: event graph works, judgment does not
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 17, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 981e294
The CLI machinery is real: soul-init built the vault, event-add recorded admiralty-rated events, and event-timeline and identity-resolve worked from bare python3. But the flagship K7 judgment engine returned a default 0.5 'derived from 0 events' after two linked events about the same person — a plain no-skill answer read the same evidence better. The advertised zero-dependency memory index errors until you pip install zvec, and its lite 'semantic' backend scored the on-topic event 0.0 for the query. validate also fails on the skill's own freshly initialized vault (missing shell/modules/_registry.md).
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 3/5
- Output vs. baseline 5/10
- Docs & honesty 2/5
What OpenTSC does
A local, offline-first 'interpersonal intelligence' system: a Python CLI that builds a markdown vault of people, admiralty-rated append-only events, and derived attribute judgments. Triggers when you ask to set up or operate an OpenTSC vault or track evidence-based judgments about collaborators. The event graph and identity tools run on bare python3, but the headline judgment engine and semantic index underdeliver in practice.
How to install OpenTSC
git clone https://github.com/opentsc/opentsc
mkdir -p ~/.claude/skills
cp -r opentsc/skill ~/.claude/skills/opentsc
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger OpenTSC
-
/opentscLocal CLI vault for contact intelligence: event graph works, judgment does not
It also activates on plain-language prompts like these:
-
Set up a vault to track evidence about how reliable my coworkers are -
Log that Bob delivered early this week, add it to his record -
How reliable is Bob based on the events I've logged about him?
Frequently asked questions
- Is the OpenTSC skill free?
- Yes. The skill itself is free from opentsc/opentsc. SkillProof publishes the install command and an independent test verdict at no cost.
- Does OpenTSC work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 17, 2026. Verdict: Works with setup. The CLI machinery is real: soul-init built the vault, event-add recorded admiralty-rated events, and event-timeline and identity-resolve worked from bare python3. But the flagship K7 judgment engine returned a default 0.5 'derived from 0 events' after two linked events about the same person — a plain no-skill answer read the same evidence better. The advertised zero-dependency memory index errors until you pip install zvec, and its lite 'semantic' backend scored the on-topic event 0.0 for the query. validate also fails on the skill's own freshly initialized vault (missing shell/modules/_registry.md).
- What is the OpenTSC SkillProof Score?
- 6.0/10 — installs cleanly 5/5, triggers reliably 3/5, output vs. baseline 5/10, docs & honesty 2/5.
- How do I install OpenTSC?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use OpenTSC with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.