Innovate, disrupt, code the future. 💻🔐🚀

A directory named agents passes plugin validate without opening the file

claude plugin validate printed Validation passed and exited 0. The directory it was given contained one file, and that file was not an agent. The same bytes, one directory higher, produced a line that named the file and a warning that the file had no front matter.

The difference is the path you pass. On Claude Code 2.1.308 the command help says it validates a plugin or marketplace manifest, or the skills, agents, and commands in a directory. A directory whose last component is the lowercase word agents takes a shorter path. The process reports success, the JSON contents array is empty, and the file is not opened. Pass the plugin root and the same file is listed as an agent.

2026-11-16

Grok plugin list cannot see the user plugin inspect calls enabled

grok plugin list printed No plugins installed and exited 0. In the same home, grok inspect printed gauge (user, enabled) and a skill count of 1. The Skills section above that row said None. grok plugin enable gauge and grok plugin details gauge both exited 1 and told the operator to run list.

That split is the contract in Grok 1.0.67, build 2fe064390d07, measured on 15 November 2026 on Linux x86_64. The user guide shipped inside that binary says Grok loads plugins from ~/.grok/plugins/, trusts that directory automatically, and leaves plugins off until the name is listed under [plugins] enabled. The public plugins page, fetched the same evening, lists the same directory and does not mention an install registry. Neither page says that inspect will mark a copied folder enabled while the Skills section stays empty, or that list, enable, and details will not open that folder.

2026-11-15

A cron notepad set by name is not the one the job reads

The job had just been renamed. Listing the notepad under the new name said it was empty. The prompt string the scheduler builds for that same job still opened its scratchpad with cursor 17, and listing the old name still returned cursor 41. Nothing in hermes cron doctor mentioned either row. The notepad command had accepted every string it was given, and the run path reads only one of them.

2026-11-15

Grant Codex project trust before a headless exec

Two isolated homes saw the same checkout. The checkout held a .codex/config.toml that named a model the public catalog does not ship, and an MCP server whose command appends one line to a file outside the checkout. On Codex CLI 0.163.0, codex exec --sandbox workspace-write loaded that model name, started that command, and wrote trust_level = "trusted" for the checkout. On 0.169.0, the same flags printed a warning, left the model name out of the event stream, did not start the command, and did not create config.toml. The later login failed in both homes. The failure is not the difference. The difference is what ran before it.

2026-11-14

A BrowseSafe++ false covers the prefix, not the page

A 6,011-character page came back as one field. The stand-in prediction was {"injected": False}. The marker that started at character 6,001 was not in the text the decision model scored. A failed call with the failure policy set to closed produced the same object, while a different counter moved. If you store only that object, you cannot tell a short clean page from a long page whose tail was never read, and you cannot tell an act-band hit from a decision that never returned a probability.

2026-11-14

Hermes plugin search misses the live catalog and stays silent

hermes plugins search --json 0xarchive exited 0. Stderr was empty. The JSON object had query set to 0xarchive and results set to an empty array. The published catalog, generated at 2026-11-14T03:45:01.238695+00:00, contains that name at version 1.0.0, pinned to 65575eddfa9c33ec429626542455eecd423cf6a5. The decoded file that holds the name is 4,443,848 bytes. The loader in this binary refuses a live body larger than 2,097,152 bytes, logs the refusal at debug, and does not write a cache. Search then answers from the catalog shipped in the checkout.

2026-11-13

Grok inspect points a project MCP server at the user config

grok mcp list printed the project server as enabled. grok inspect --json named the same server and set source.path to the user config. That file did not contain the server. grok mcp doctor refused to start it, and the hint said to re-run with --trust. That flag is not in grok --help, and putting it after doctor is an error.

The file a script will open is the file inspect names. On this binary that file is the wrong one for a project server, before the folder is trusted and after the grant is written.

2026-11-13

An engines range does not unload a Claude Code 2.1.308 plugin

Claude Code 2.1.323, in the changelog dated 12 November 2026, adds engines["claude-code"] to plugin.json and says a plugin outside that range is not installed, updated, or loaded. The binary in this lab is 2.1.308. On that binary the same key is a warning, the warning says the field is ignored, and a plugin whose range sits below 2.0.0 still seats its skill.

That is a version boundary, not a disagreement inside one release. Copying the new field onto a fleet that has not upgraded does not do the job the changelog describes. The validator on 2.1.308 will tell you it is safe to keep the field. That sentence matches what this binary did. It is not a description of 2.1.323, and it is not a reason to skip the list after you upgrade.

2026-11-12

A failed kanban completion still binds the pull request

The card was still blocked. hermes kanban show --json listed completion_contract as https://github.com/acme/repo/pull/7. Create, a few seconds earlier, had stored acme/repo. Complete had exited 1. Stderr was one sentence: cannot complete t_ab9031f6 (unknown id or terminal state). The id was the one create had printed. The status was not terminal. No gh process had been started for that card.

The kanban page already says which login must read a pull request. It does not say that a failed read rewrites the stored contract, or that complete reports the refusal as a missing id. Those two results are what this binary recorded.

2026-11-12

Read a Codex feature stage before you copy its boolean

Codex CLI 0.163.0 will print every feature it knows as a key, a stage, and a boolean. The config page tells you to turn a feature off by setting that key to false. The page is describing the file. The process reads a resolved set. On an empty home, eight rows whose stage is removed already have the boolean true. Writing false moved two removed keys this run measured and left three others at true. codex features disable exited 0 for one of the keys that did not move, wrote the file, and printed a success line. The next list still said true.

2026-11-11