amazon-quick-exec-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| QUICK_EXEC_SHELL | No | Which shell to use | /bin/zsh |
| QUICK_EXEC_AUDIT_LOG | No | Audit log path | ~/.quick-exec-mcp/audit.jsonl |
| QUICK_EXEC_LOG_LEVEL | No | INFO logs every call to stderr | WARNING |
| QUICK_EXEC_MAX_BYTES | No | Per-stream output cap, roughly 15k tokens | 60000 |
| QUICK_EXEC_SHELL_ARGS | No | -l sources the profile; drop it to skip that | -lc |
| QUICK_EXEC_DEFAULT_CWD | No | Working directory when cwd is omitted | $HOME |
| QUICK_EXEC_MAX_TIMEOUT | No | Ceiling on the timeout argument | 3600 |
| QUICK_EXEC_READER_GRACE | No | Seconds to keep draining the pipes after the child exits | 2 |
| QUICK_EXEC_ALLOW_DANGEROUS | No | 1 turns the command guard off | |
| QUICK_EXEC_DEFAULT_TIMEOUT | No | Default timeout, seconds | 120 |
| QUICK_EXEC_MAX_FINISHED_JOBS | No | Finished jobs retained before the oldest are dropped | 50 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| shell_executeA | Run a shell command on this Mac and wait for it to finish. Runs through a login shell, so pipes, globs, redirects, &&, $VARS, heredocs and shell functions all work — pass the command exactly as you would type it in Terminal. Use this for anything that finishes within the timeout. For work that outlives the timeout (builds, installs, dev servers) use shell_start instead. Args: command: Shell command line, e.g. "ls -la ~/Documents | head -20". cwd: Working directory. Defaults to the user's home. ~ is expanded. timeout: Seconds before the process tree is killed. Default 120, max 3600. Zero or negative means "use the default", not "give up immediately". stdin: Text piped to the command's stdin. env: Extra environment variables, merged over the inherited environment. Returns: exit_code, stdout, stderr, duration_s, cwd, timed_out, and stdout_truncated / stderr_truncated flags. |
| shell_startA | Start a long-running command in the background and return a job id. Output is buffered as it arrives; read it with shell_job_output, block on it with shell_job_wait, stop it with shell_job_kill. Use this for builds, installs, test suites and dev servers that would blow shell_execute's timeout. Args: command: Shell command line. cwd: Working directory. Defaults to the user's home. env: Extra environment variables. Returns: job_id, pid, command, cwd. |
| shell_job_outputA | Read the output captured so far from a background job. Safe to call repeatedly while the job runs — it returns a snapshot and does not consume the buffer. Args: job_id: Id returned by shell_start. tail_lines: Return only the last N lines of each stream. 0 means all. Returns: running, exit_code (null while running), stdout, stderr, elapsed_s. |
| shell_job_waitA | Block until a background job exits, then return its output. Cheaper and more reliable than polling shell_job_output in a loop. If the
timeout expires the job keeps running and Args: job_id: Id returned by shell_start. timeout: Max seconds to wait. The job is NOT killed on timeout. tail_lines: Return only the last N lines of each stream. 0 means all. |
| shell_job_killA | Kill a background job and its whole process tree. Args: job_id: Id returned by shell_start. |
| shell_list_jobsB | List background jobs started this session, running and finished. |
| shell_infoA | Report how this server is configured and what environment commands see. Worth calling first in a session: it proves the server is running on the real host as the real user rather than inside Quick's sandbox. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |