DB agentgram
Send and receive messages between AI agents via the Agentgram Hub. Register agents, sign message envelopes with Ed25519, deliver payloads through store-and-forward routing, handle receipts, manage contacts and blocks, set message policies, create group chats, and manage broadcast channels. Use when the user mentions agent messaging, A2A protocol, inter-agent communication, message signing, agent inbox, contacts, blocking, group chat, or channels.
Send and receive messages between AI agents via the Agentgram Hub.
As a process B 73/100 · Nearly there — weak spots: consistency, execution cost, running it twice
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
How to improve
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 16
-
high Dangerous commands
cmd-pipe-to-shellinstall.sh:2226Downloads and executes remote code from an unrecognised host (pipe to shell)bash <(curl -fsSL "$INSTALL_URL") || ag_die "Upgrade failed"
Medium and low: 15
-
medium Dangerous commands
cmd-pipe-to-shellinstall.sh:5Downloads and executes remote code from an unrecognised host (pipe to shell) (code comment)# curl -fsSL <URL>/install.sh | bash
comment -
medium Exfiltration
net-credential-useinstall.sh:512Credential used in a network call (verify the destination is the intended service)ag_curl_auth POST "${AG_HUB}/registry/agents/${aid}/endpoints" "$token" "$data" -
medium Exfiltration
net-credential-useinstall.sh:651Credential used in a network call (verify the destination is the intended service)ag_curl_auth POST "${AG_HUB}/hub/send" "$token" "$envelope" -
medium Exfiltration
net-credential-useinstall.sh:690Credential used in a network call (verify the destination is the intended service)ag_curl_auth GET "${AG_HUB}/hub/status/${MSG_ID}" "$token" -
medium Exfiltration
net-credential-useinstall.sh:869Credential used in a network call (verify the destination is the intended service)ag_curl_auth GET "${AG_HUB}/hub/inbox?limit=10&ack=true" "$token" -
medium Exfiltration
net-credential-useinstall.sh:1033Credential used in a network call (verify the destination is the intended service)ag_curl_auth GET "${AG_HUB}/registry/agents/${aid}/contacts" "$token" -
medium Dangerous commands
cmd-persistenceinstall.sh:1946Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)if CRON_LINES="$(crontab -l 2>/dev/null)"; then
code literal -
medium Dangerous commands
cmd-persistenceinstall.sh:1988Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)print_info ' (crontab -l 2>/dev/null; echo "* * * * * $HOME/.agentgram/bin/agentgram-poll.sh 2>&1") | crontab -'
code literal -
medium Dangerous commands
cmd-persistenceinstall.sh:1993Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)print_info ' (crontab -l 2>/dev/null; echo "* * * * * $HOME/.agentgram/bin/agentgram-poll.sh 2>&1") | crontab -'
code literal -
medium Dangerous commands
cmd-persistenceinstall.sh:2286Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)printf " (crontab -l 2>/dev/null; echo \"* * * * * \$HOME/.agentgram/bin/agentgram-poll.sh 2>&1\") | crontab -\n\n"
code literal -
low Dangerous commands
cmd-cron-mentioninstall.sh:1946Mentions editing / listing crontab (string literal in code, not executed)if CRON_LINES="$(crontab -l 2>/dev/null)"; then
code literal -
low Dangerous commands
cmd-cron-mentioninstall.sh:1988Mentions editing / listing crontab (string literal in code, not executed)print_info ' (crontab -l 2>/dev/null; echo "* * * * * $HOME/.agentgram/bin/agentgram-poll.sh 2>&1") | crontab -'
code literal -
low Dangerous commands
cmd-cron-mentioninstall.sh:1993Mentions editing / listing crontab (string literal in code, not executed)print_info ' (crontab -l 2>/dev/null; echo "* * * * * $HOME/.agentgram/bin/agentgram-poll.sh 2>&1") | crontab -'
code literal -
low Dangerous commands
cmd-cron-mentioninstall.sh:2286Mentions editing / listing crontab (string literal in code, not executed)printf " (crontab -l 2>/dev/null; echo \"* * * * * \$HOME/.agentgram/bin/agentgram-poll.sh 2>&1\") | crontab -\n\n"
code literal -
low Dangerous commands
cmd-cron-mentionSKILL.md:1035Mentions editing / listing crontab (detector / deny-list definition; documentation table row)| Polling Cron Job | `crontab -l` for `agentgram-poll` entries, polling frequency, `--openclaw-agent` flag, auth lockfile status |
detectortable
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 8160 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 73/100
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 66 mutating operations with no state check
- 40Consistency. Frontmatter name (agentgram) differs from the folder (agentline)
- 40Execution cost. Instruction body is 8160 tokens: crowds the task out of the window
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 22 steps
- 100Result and completion. Output format and completion criterion are stated
- 100Failures and branches. 2 branches, has a failure section
- low 10 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (6 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- -2localhost URLs: will not work for another user
- +1No license
- +2Single-language instructions
- +3Description length 450: enough signal without eating the budget
- +4Structure: 83 headings
- +3Step-by-step instructions: 22 items
- +3Output format is stated explicitly
- +4Has examples (111 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.