{"slug": "edit-one-line-of-an-ai-agent-journal-and-see-what-aisync-verify-says", "title": "Edit one line of an AI agent journal and see what aisync verify says", "summary": "The maintainer of aisync, a local CLI for multi-agent context management and observability across AI coding tools, demonstrated that its hash-chained event journal detects hand edits: modifying an event in journal.jsonl made `aisync verify` report \"line 4: id mismatch (event edited by hand)\" with exit code 1, and deleting a line produced \"line 2: prev link broken\" with exit code 1, while the untouched journal verified as intact. The maintainer cautioned that the chain catches accidental or careless changes but is not a signature and should not be called tamper-proof, since an attacker recomputing later hashes was not tested.", "body_md": "aisync is a local CLI for Multi-Agent Context Management, Coordination & Observability across AI coding tools. I maintain it. Part of the hub is a journal that every agent event goes into, and the journal is hash-chained. This post tests that claim by editing the journal by hand and seeing what `aisync verify` says.\n\nI used a throwaway virtualenv and pointed `AISYNC_HOME` at a temp directory, so my real hub was never touched. Version 0.2.9. The PyPI package is called `aisync-cli`, not `aisync`.\n\n```\npython3 -m venv v && ./v/bin/pip install aisync-cli\nexport AISYNC_HOME=\"$PWD/home\"\n./v/bin/aisync init\nfor i in 1 2 3; do ./v/bin/aisync log --agent claude-code --project demo --kind dev_log --summary \"step $i\"; done\n./v/bin/aisync verify\n```\n\nThat gave a journal of five lines and `journal chain intact`, exit code 0.\n\nI changed `step 2` to `step 2 (edited)` directly in `journal.jsonl` and ran verify again:\n\n```\nline 4: id mismatch (event edited by hand)\nexit=1\n```\n\nI restored the file, deleted its second line and ran verify:\n\n```\nline 2: prev link broken (event lost, reordered or forked)\nexit=1\n```\n\nWith the original file back, verify printed `journal chain intact` and exited 0.\n\nWhen Claude Code, Codex and Cursor all write to one journal, the question after a bad day is \"which agent did what, and has anyone changed the record since\". Every event stores the id of the one before it. An edit changes that event's id. A removal breaks the link to the next event. Both show up as a non-zero exit, so a script or a CI step can check it.\n\nI did not test an attacker who recomputes every later hash after editing. A hash chain on its own detects accidental or careless changes. It is not a signature, so I would not call the journal tamper-proof. Only the three cases above were run.\n\n`pip install aisync-cli`\nThis post and the test run were prepared with AI assistance (Claude Code). The outputs above are copied from that run.", "url": "https://wpnews.pro/news/edit-one-line-of-an-ai-agent-journal-and-see-what-aisync-verify-says", "canonical_source": "https://dev.to/maxont/edit-one-line-of-an-ai-agent-journal-and-see-what-aisync-verify-says-1202", "published_at": "2026-11-16 09:47:50+00:00", "updated_at": "2026-11-16 09:51:03.405425+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools", "mlops"], "entities": ["aisync", "aisync-cli", "Claude Code", "Codex", "Cursor", "PyPI"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/edit-one-line-of-an-ai-agent-journal-and-see-what-aisync-verify-says", "markdown": "https://wpnews.pro/news/edit-one-line-of-an-ai-agent-journal-and-see-what-aisync-verify-says.md", "text": "https://wpnews.pro/news/edit-one-line-of-an-ai-agent-journal-and-see-what-aisync-verify-says.txt", "jsonld": "https://wpnews.pro/news/edit-one-line-of-an-ai-agent-journal-and-see-what-aisync-verify-says.jsonld"}}