A command for changing the scope of an active issue after its specification turns out to be incomplete or incorrect.

It re-reads the issue, updates allowed and disallowed files and required checks, clears verification and review receipts, and records why the change was made.

It keeps the recorded issue scope aligned with the updated specification and makes previously verified work undergo verification again.

issue-amend — Command by assafkip

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/assafkip/kipi-system/issue-amend

Clone the repo

git clone --depth 1 https://github.com/assafkip/kipi-system

Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 601 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.

Model	Per session	Once invoked
Fable 5	$0.00026	$0.00601
Opus 5	$0.00013	$0.00300
Sonnet 5	$0.00005	$0.00120
Haiku 4.5	$0.00003	$0.00060

Measured yesterday against content hash 76870a6e5b55, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

issue-amend 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 yesterday.

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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

issue-amend — 97% identical, 2 lines differ

plugins/kipi-dsse/commands/issue-amend.md · 39 lines

What it actually says

Use this when you realize mid-build that the spec was wrong. The flow:

The founder edits the issue spec in q-ktlyst/.q-system/issues/<id>.md (or the instance-equivalent path). Typical edits: widen allowed_files, adjust required_checks, add a new disallowed_files entry.
Run amend with a short reason explaining what changed and why:
```
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/issue_runner.py" amend \
  --reason "realized the bug also lives in helpers/*, widening scope"
```
The runner does four things:
- Re-reads the spec from disk.
- Re-snapshots allowed_files, required_checks, and disallowed_files into state.
- Clears receipts.verified and receipts.reviewed (forces re-verify + re-review).
- Appends an entry to state.amendments with timestamp, reason, old snapshot, new snapshot.
What amend does NOT change:
- review_rounds counter (the cap still governs; amending does not reset review budget).
- Spec status (stays in-progress).
- findings_triaged receipt (that receipt only matters at closeout).
Exit 0 means the amendment is logged. Exit 2 means either no active issue or missing --reason.
Re-run /issue-verify to regenerate the verified receipt against the new scope. Then /issue-review for the reviewed receipt. The runner's review-rounds cap still applies; amending does not grant extra review budget. If a full standard + adversarial pair already ran pre-amend, ISSUE_ALLOW_REVIEW_REPEAT=1 is the override knob.
On /issue-closeout, close flushes all entries from state.amendments into an ## Amendments section at the bottom of the spec file. That becomes the permanent audit record. After close, the spec shows exactly what was amended, when, why, and what the scope looked like before and after each amendment.

Guardrails:

Amend is for genuine mid-build spec corrections. It is not a retry button after a bad review. If the issue as specified was fine but the implementation was wrong, fix the implementation. Do not amend.
Amend requires a non-empty --reason. Empty strings are rejected.
Amend does not create or modify files other than .claude/state/active-issue.json. The spec edit is a separate manual step before amend.
Repeated amendments are fine. Each one records its own before/after snapshot and adds its own entry to the permanent footer on close.

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.

yesterday First seen · 39 lines · 26 tokens per session scan A 76870a6e5b55

Subscribe to this mod's changes

issue-amend is a command published in the GitHub repository assafkip/kipi-system (108 stars, last pushed today), licensed MIT. It adds 26 tokens to every session and 601 once invoked, about $0.0001 per session on Opus 5. 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.

Planning agent-framework ai-agents ai-memory claude-code knowledge-management markdown

issue-amend

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

Copies of this mod

What it actually says

What this file has done since we first saw it

Other commands, from other repositories

capture

recall

custodian

integrate

query

search