- Home
- Skills
- Technical Documentation
- Continuity Record
Works with the AI Agents you already use
Continuity Record
Preserve the objective, current phase, authority, decisions, evidence, blockers, and next bounded action in a compact handoff that survives context or executor changes.
$9
Continuity Record
Example session with this skill installed
Create a continuation record before handing this repository task to another executor: the auth migration is partially complete, the approved JWT decisions are settled, tests for login and refresh pass, middleware still reads the old session, and the revocation store is undecided.
- Read your context and instructions
- Compiled the continuity record
Illustrative worked example. Synthetic data from the package; this is not an observed external execution.
OBJECTIVE: migrate auth from sessions to JWT.
CURRENT PHASE: implementation (2 of 4 endpoints migrated).
AUTHORITY: ADR-014 (JWT, 15-min access + refresh); glossary term "principal".
DECISIONS: RS256; refresh token in httpOnly cookie; access token in memory.
CHANGED SURFACES: auth/jwt.js (new), routes/login.js, routes/refresh.js (done); routes/logout.js, middleware/auth.js (pending).
EVIDENCE / CHECKS: login+refresh tests green; middleware still reads old session - expected until migrated.
BLOCKERS / UNKNOWNS: logout must revoke refresh token - revocation store not yet chosen (UNVERIFIED).
NEXT BOUNDED ACTION: implement revocation list (decide store first), then migrate middleware/auth.js.
DO NOT REPEAT: re-deriving the RS256 vs HS256 choice - settled in ADR-014.
Connects securely to your tools. The creator never sees your data.
About this skill
When context, executor, session, or ownership is about to change, Continuity Record preserves operational state instead of replaying the conversation. It recovers the current objective and phase, authoritative artifacts, accepted decisions, changed surfaces, checks, blockers and next action, then prunes disposable history into a compact handoff.
What it does
It keeps the authority order explicit and reads the smallest sufficient evidence. It preserves decisions that prevent re-litigation, marks uncertainty instead of hiding it, and keeps the record focused on the next bounded action.
How it works
- Recover the current objective, phase, authority, decisions, changed surfaces, checks, blockers and next bounded action.
- Verify facts from the repository or runtime when that is cheap, without assuming shell, Git, browser, network or other capabilities.
- Prune long logs, obsolete hypotheses, repeated discussion and facts that are cheap to rediscover.
- Write the prescribed continuation record when an approved handoff location and write authority exist; otherwise return the record without inventing repository governance.
Output
The handoff follows a fixed structure:
- OBJECTIVE
- CURRENT PHASE
- AUTHORITY
- DECISIONS
- CHANGED SURFACES
- EVIDENCE / CHECKS
- BLOCKERS / UNKNOWNS
- NEXT BOUNDED ACTION
- DO NOT REPEAT
Boundaries
This is a continuity checkpoint, not a conversation log, project tracker, implementation workflow, or release verdict. It does not include secrets or credentials, mutate application code, install dependencies, commit, push, merge, deploy, or run destructive migrations. It writes only the continuity artifact when an approved location and write authority exist.
Why it is different
A conversation summary preserves discussion, and a task tracker preserves work items. Continuity Record preserves the operational state a fresh executor needs to resume: authority, accepted decisions, changed surfaces, exact checks, blockers, uncertainty, next action and actions not to repeat.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Technical Documentation.
This is a continuity checkpoint, not a conversation log, task tracker, implementation workflow, or release verdict. It requires enough current-state evidence to identify the objective, phase, authority, decisions, changed surfaces, checks, blockers, and next action. It does not include secrets or credentials or modify application code. Writing requires an approved handoff location and write authority; otherwise the record is returned. Shell, Git, browser, network, MCP, and package-manager capabilities are not assumed; unavailable checks are marked UNVERIFIED or BLOCKED.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 17 days ago
Permissions
Requires read access to current task state and authoritative artifacts. Writing the continuity artifact is optional when an approved handoff location and write authority exist; otherwise return the record without inventing repository governance. Terminal/Shell is optional for focused fact checks with repository-native or manual fallback. Browser, network access, MCP services, external accounts and environment variables are not required by the package.