How do I install the Precedent skill?

Run npx skillmds@latest add tonydzi/precedent in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Precedent safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Precedent?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Precedent free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under MIT.

Who published Precedent?

tonydzi (@tonydzi) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Precedent by tonydzi — Agent Skill

Name: Precedent
Author: tonydzi

Precedent

Check whether something was already decided before proposing it: searches the decisions journal, the declined registry and the rules codex plus semantic search, and returns what was decided, why, what was rejected, and under what condition to revisit. Triggers: "/precedent <topic>", "did we already decide this".

tonydzi 7bf32f9 3.8 KB Updated Sep 17, 2026

File contents

/precedent — have we decided this already? (find the precedent before deciding)

Cheap insurance BEFORE a new structural choice or proposal: maybe there is already a verdict on this topic (accepted / rejected / deferred) — so we do not re-pitch something already declined, nor re-decide what is decided. This is the "RECALL before activity" rule applied to the DECISION layer.

What to pull up (cheap → expensive; stop as soon as you find it)

The declined journal (first — the most frequent hit): memory $USERPROFILE/.claude/projects/<project>/memory/declined-decisions.md — what was rejected or deferred, why in the owner's own words, and Revisit-if (the condition under which it can come back).
The accepted-decisions journal (vault): grep $OBSIDIAN_VAULT/02-Decisions/ (subfolders by domain) and any decision-* files in the vault.
```
grep -rinl "<topic/keywords>" "$OBSIDIAN_VAULT/02-Decisions"
```
The Bible (rules for all actors): grep reglament-* / protocol-* in the vault — maybe a standing rule already closes the question.
Semantic top-up (if the exact grep is empty but the topic was clearly discussed): the /ask skill (RAG over the vault) — it catches phrasings grep missed.

What to give back to the operator

A short verdict, not a dump:

A precedent EXISTS → what was decided · when · why · a link to the note. If it was REJECTED, name the Revisit-if: reopen ONLY if the condition is met, or explicitly as a trade-off ("there is no other way") — never silently re-pitch.
No precedent → say exactly that: "I found no decision on this" (after checking the RIGHT drive/folder — an empty result is often "looked in the wrong place", not "no data"), and then go decide from scratch (for strategic questions, via the Alpha Protocol R+DR).

When to call it

Before I propose a new field / script / structure / rule / automation (per show-before-after).
When the operator says "I think we discussed this".
At the start of the Alpha Protocol — as part of the RECALL step.

Boundaries

Read-only: it only surfaces the past, it decides nothing for the operator.
It does not duplicate the Alpha Protocol — it is that protocol's narrow sub-step "is there already a verdict"; a full strategy still needs R+DR.

Like this skill? It is one of 100 in second-brain-starter-kit: the second brain we built for ourselves and run every day at Palo Alto AI Research Lab. Install the whole set with npx skills add tonydzi/second-brain-starter-kit. Everything is open source and free, so take what you need.

Flagships worth a look on their own: secondop-panel (a second opinion from a panel of external models), claude-memory-tidy (stop your agent's memory from rotting), telegram-mcp-kit (your own Telegram over MCP in about 15 minutes).

Author: Anton Dziatkovskii, Palo Alto AI Research Lab. Telegram @tonydzi - WhatsApp +1 341 222 9178 - X @Tony_Stef_

Engineers: want to test-drive this setup? Message me. I hand out free starter seeds to engineers who test and report back, and custom skill requests are welcome.

---name: precedentdescription: Check whether something was already decided before proposing it: searches the decisions journal, the declined registry and the rules codex plus semantic search, and returns what was decided, why, what was rejected, and under what condition to revisit. Triggers: "/precedent <topic>", "did we already decide this".license: MIT---# /precedent — have we decided this already? (find the precedent before deciding)Cheap insurance BEFORE a new structural choice or proposal: maybe there is already a verdict on this topic (accepted / rejected / deferred) — so we do not re-pitch something already declined, nor re-decide what is decided. This is the "RECALL before activity" rule applied to the DECISION layer.## What to pull up (cheap → expensive; stop as soon as you find it)1. **The declined journal (first — the most frequent hit):** memory `$USERPROFILE/.claude/projects/<project>/memory/declined-decisions.md` — what was rejected or deferred, why in the owner's own words, and `Revisit-if` (the condition under which it can come back).2. **The accepted-decisions journal (vault):** grep `$OBSIDIAN_VAULT/02-Decisions/` (subfolders by domain) and any `decision-*` files in the vault. ```bash grep -rinl "<topic/keywords>" "$OBSIDIAN_VAULT/02-Decisions" ```3. **The Bible (rules for all actors):** grep `reglament-*` / `protocol-*` in the vault — maybe a standing rule already closes the question.4. **Semantic top-up (if the exact grep is empty but the topic was clearly discussed):** the `/ask` skill (RAG over the vault) — it catches phrasings grep missed.## What to give back to the operatorA short verdict, not a dump:- **A precedent EXISTS** → what was decided · when · why · a link to the note. If it was REJECTED, name the `Revisit-if`: reopen ONLY if the condition is met, or explicitly as a trade-off ("there is no other way") — never silently re-pitch.- **No precedent** → say exactly that: "I found no decision on this" (after checking the RIGHT drive/folder — an empty result is often "looked in the wrong place", not "no data"), and then go decide from scratch (for strategic questions, via the Alpha Protocol `R+DR`).## When to call it- Before I propose a new field / script / structure / rule / automation (per `show-before-after`).- When the operator says "I think we discussed this".- At the start of the Alpha Protocol — as part of the RECALL step.## Boundaries- Read-only: it only surfaces the past, it decides nothing for the operator.- It does not duplicate the Alpha Protocol — it is that protocol's narrow sub-step "is there already a verdict"; a full strategy still needs `R+DR`.------**Like this skill?** It is one of 100 in [second-brain-starter-kit](https://github.com/tonydzi/second-brain-starter-kit): the second brain we built for ourselves and run every day at Palo Alto AI Research Lab. Install the whole set with `npx skills add tonydzi/second-brain-starter-kit`. Everything is open source and free, so take what you need.Flagships worth a look on their own: [secondop-panel](https://github.com/tonydzi/secondop-panel) (a second opinion from a panel of external models), [claude-memory-tidy](https://github.com/tonydzi/claude-memory-tidy) (stop your agent's memory from rotting), [telegram-mcp-kit](https://github.com/tonydzi/telegram-mcp-kit) (your own Telegram over MCP in about 15 minutes).Author: **Anton Dziatkovskii**, Palo Alto AI Research Lab. Telegram [@tonydzi](https://t.me/tonydzi) - WhatsApp [+1 341 222 9178](https://wa.me/13412229178) - X [@Tony_Stef_](https://x.com/Tony_Stef_)**Engineers: want to test-drive this setup?** Message me. I hand out free starter seeds to engineers who test and report back, and custom skill requests are welcome.

Frequently asked questions