topic

A command for checking whether text has shifted from the project currently associated with a session to another project.

In plain words
What is it for?
Use it to test a piece of context, set the current project and confidence threshold, return JSON, and use its exit code in automated checks.
Why use it?
It can flag possible project mix-ups before work is routed or recorded in the wrong place.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/mnott/pai/topic
Clone the repo
git clone --depth 1 https://github.com/mnott/PAI
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 492 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00492
Opus 5 $0.00000 $0.00246
Sonnet 5 $0.00000 $0.00098
Haiku 4.5 $0.00000 $0.00049

Measured 2d ago against content hash 2c7a62f7ebbc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

topic scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

docs/commands/topic.md · 47 lines

What it actually says

pai topic

Topic shift detection: check whether context has drifted to a different project

Synopsis

pai topic <subcommand> [options]

Subcommands

Command Description
pai topic check <context> Check whether context text has drifted to a different project.

pai topic check

Check whether context text has drifted to a different project.

Example: pai topic check "working on JWT authentication" Exit code 1 if a shift is detected (useful for hooks).

Arguments

Argument Kind
<context> required

Options

Option Description Default
--current <slug> The project this session is currently routed to
--threshold <n> Confidence threshold [0-1] to declare a shift (default: 0.6)
--json Output raw JSON result

See also

pai backup · pai clear-names · pai daemon · pai db · pai end · pai help · pai identity · pai kg · pai mcp · pai memory · pai notify · pai observation · pai obsidian · pai pause · pai project · pai projects · pai registry · pai restore · pai session · pai sessions · pai setup · pai shell-init · pai skill · pai task · pai update · pai zettel

Run pai help <area> to read any of these in the terminal.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 47 lines · 0 tokens per session scan A 2c7a62f7ebbc

Subscribe to this mod's changes

topic is a command published in the GitHub repository mnott/PAI (45 stars, last pushed 8d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 492 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.