signal-state
Continuous disk-backed state engine for the SIGNAL protocol
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Menu
Continuous disk-backed state engine for the SIGNAL protocol
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Manually triggers a checkpoint compression of the current session state. Collapses conversation history into a compact state atom (≤50 tokens). Use when user types /signal-ckpt, "checkpoint", "compress context", "summarize session", "save state", or when the context window is getting large. In SIGNAL-3 mode this fires automatically every 5 turns.
Stage all changes and create a conventional commit from the diff. Zero prompts, zero confirmation by default. Use when user types /signal-commit, "commit everything", "just commit", "signal commit", or asks to commit current changes without specifying a message. Supports --draft (show message without committing) and --split (atomic commits per logical change).
Compress memory files, prompts, rules, handoff notes, and project docs into SIGNAL-shaped prompt-token-friendly text while preserving code, paths, identifiers, errors, versions, and technical terms. Use when user types /signal-compress, "compress this memory", "shrink this prompt", "compress project rules", or asks to reduce input/prompt tokens.
Stage all changes, commit, push, then open a pull request. Title and body are generated from the diff and commit history. Use when user types /signal-pr, "open a PR", "create pull request", "ship a PR", or "make a PR". Requires gh CLI.
Stage all changes, create a conventional commit, then push to remote. One command for the full local→remote cycle. Use when user types /signal-push, "commit and push", "push everything", "push my changes", or wants to send current changes to remote without manual steps.
Review code in SIGNAL's one-line format with required severity per issue. Output is typed, structured, and chainable — not prose paragraphs. Use when user types /signal-review, "review this", "review my code", "check this PR", "review this file", or asks for a code review on any file, diff, or PR. Works standalone or as part of a SIGNAL session.
| name | signal-state |
| description | Continuous disk-backed state engine for the SIGNAL protocol |
| keywords | ["signal","state","memory","checkpoint","status"] |
This skill manages the Continuous State Engine for the SIGNAL protocol. It transitions the context window from an ephemeral chat history to a persistent .signal_state.md file on disk.
.signal_state.md if it exists. It IS your context./signal-state → refresh file immediately.Use write_file or replace to modify .signal_state.md in project root. No scripts required.
.signal_state.md)YAML frontmatter + 3 blocks: Context, Progress, Next.
---
signal_version: 0.2
boot_mode: BOOT:arch
updated_at: 2026-04-15T15:00:00Z
---
## Context
project=move-validator
stack=node+ts
scope=knight-validation
## Progress
[x] parse files
[x] validate L-shape
[x] validate off-board edge cases
[/] SAN to coord conversion boundary
[ ] write unit tests for SAN parse
## Next
turn_focus=implement SAN string parsing in src/move.ts
[ ] to [x]. Do not just append completed items to an infinitely growing list.Progress list exceeds 10 items, collapse the completed items [x] into a single summary line: [x] core knight validation (5 tasks).If the user asks to start using continuous state and the file doesn't exist, create it immediately based on the current context of the chat.
# Example action
write_to_file ".signal_state.md" <formatted_state_content>